@cursor/july 0.1.111 → 0.1.113
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +4 -0
- package/dist/bin/agent-serve.js +20 -4
- package/dist/channels/checks.d.ts +10 -0
- package/dist/channels/checks.d.ts.map +1 -1
- package/dist/channels/github/github-channel.d.ts +29 -19
- package/dist/channels/github/github-channel.d.ts.map +1 -1
- package/dist/channels/github/github-channel.js +29 -19
- package/dist/channels/origin/checks.d.ts +1 -1
- package/dist/channels/origin/checks.d.ts.map +1 -1
- package/dist/channels/slack/agentic-delivery.d.ts.map +1 -1
- package/dist/channels/slack/agentic-delivery.js +2 -0
- package/dist/channels/slack/dispatch.d.ts +11 -1
- package/dist/channels/slack/dispatch.d.ts.map +1 -1
- package/dist/channels/slack/dispatch.js +146 -39
- package/dist/channels/slack/placeholder.d.ts +26 -0
- package/dist/channels/slack/placeholder.d.ts.map +1 -1
- package/dist/channels/slack/placeholder.js +28 -0
- package/dist/channels/slack/slack-channel.d.ts.map +1 -1
- package/dist/channels/slack/slack-channel.js +9 -16
- package/dist/docs/404.html +2 -2
- package/dist/docs/assets/{app.Drol6mi6.js → app.CAeK13eM.js} +4 -4
- package/dist/docs/assets/building-with-agents.md.BBCx0AUo.js +9 -0
- package/dist/docs/assets/building-with-agents.md.BBCx0AUo.lean.js +1 -0
- package/dist/docs/assets/chunks/@localSearchIndexroot.Ck9E52Ls.js +1 -0
- package/dist/docs/assets/chunks/{VPLocalSearchBox.CdqoBZJX.js → VPLocalSearchBox.C9LbPHod.js} +1 -1
- package/dist/docs/assets/chunks/{arc.BVX3ycTn.js → arc.CmMq2zmS.js} +1 -1
- package/dist/docs/assets/chunks/{architectureDiagram-Q4EWVU46.CcChnMxX.js → architectureDiagram-Q4EWVU46.CCXB8Uj5.js} +1 -1
- package/dist/docs/assets/chunks/{baseUniq.xwtXO-yt.js → baseUniq.CyQo6eLe.js} +1 -1
- package/dist/docs/assets/chunks/{blockDiagram-DXYQGD6D.CijZ_taK.js → blockDiagram-DXYQGD6D.JYq6w91N.js} +1 -1
- package/dist/docs/assets/chunks/{c4Diagram-AHTNJAMY.0FBvwBvK.js → c4Diagram-AHTNJAMY.BRV8GPJJ.js} +1 -1
- package/dist/docs/assets/chunks/channel.BHiYmnZ4.js +1 -0
- package/dist/docs/assets/chunks/{chunk-4BX2VUAB.CELMmMDA.js → chunk-4BX2VUAB.Bv4ooYQR.js} +1 -1
- package/dist/docs/assets/chunks/{chunk-4TB4RGXK.C4bhFtSm.js → chunk-4TB4RGXK.t4JtKPcj.js} +1 -1
- package/dist/docs/assets/chunks/{chunk-55IACEB6.ZnQ9gRRQ.js → chunk-55IACEB6.34lCHj9Y.js} +1 -1
- package/dist/docs/assets/chunks/{chunk-EDXVE4YY.DPlmglG-.js → chunk-EDXVE4YY.BSwrPNrt.js} +1 -1
- package/dist/docs/assets/chunks/{chunk-FMBD7UC4.BrUk8pff.js → chunk-FMBD7UC4.Beeun-R-.js} +1 -1
- package/dist/docs/assets/chunks/{chunk-OYMX7WX6.CcmWIncu.js → chunk-OYMX7WX6.BUUFUcJc.js} +1 -1
- package/dist/docs/assets/chunks/{chunk-QZHKN3VN.D65-cs8I.js → chunk-QZHKN3VN.B2XjHzN_.js} +1 -1
- package/dist/docs/assets/chunks/{chunk-YZCP3GAM.qoXZpG9F.js → chunk-YZCP3GAM.CLYG8znk.js} +1 -1
- package/dist/docs/assets/chunks/classDiagram-6PBFFD2Q.Degh8l90.js +1 -0
- package/dist/docs/assets/chunks/classDiagram-v2-HSJHXN6E.Degh8l90.js +1 -0
- package/dist/docs/assets/chunks/clone.BIywbczV.js +1 -0
- package/dist/docs/assets/chunks/{cose-bilkent-S5V4N54A.8rYtqudO.js → cose-bilkent-S5V4N54A.DVEa6fZp.js} +1 -1
- package/dist/docs/assets/chunks/{dagre-KV5264BT.DrRP1fOh.js → dagre-KV5264BT.C9PZQK-S.js} +1 -1
- package/dist/docs/assets/chunks/{diagram-5BDNPKRD.DHj_xA2_.js → diagram-5BDNPKRD.DoN0uv3Y.js} +1 -1
- package/dist/docs/assets/chunks/{diagram-G4DWMVQ6.Bz_6nAKj.js → diagram-G4DWMVQ6.Czv3duqx.js} +1 -1
- package/dist/docs/assets/chunks/{diagram-MMDJMWI5.BWA0xSW9.js → diagram-MMDJMWI5.BinJ5kWb.js} +1 -1
- package/dist/docs/assets/chunks/{diagram-TYMM5635.CpTJLNJI.js → diagram-TYMM5635.DW326M4K.js} +1 -1
- package/dist/docs/assets/chunks/{erDiagram-SMLLAGMA.-7AWWSrP.js → erDiagram-SMLLAGMA.U2pR_OA7.js} +1 -1
- package/dist/docs/assets/chunks/{flowDiagram-DWJPFMVM.BTnQ742_.js → flowDiagram-DWJPFMVM.ByWJXeYK.js} +1 -1
- package/dist/docs/assets/chunks/framework.BNw1pucY.js +19 -0
- package/dist/docs/assets/chunks/{ganttDiagram-T4ZO3ILL.B5_HiiQ5.js → ganttDiagram-T4ZO3ILL.OquF0Rtg.js} +1 -1
- package/dist/docs/assets/chunks/{gitGraphDiagram-UUTBAWPF.CZcNTFZd.js → gitGraphDiagram-UUTBAWPF.Bpn01P7X.js} +1 -1
- package/dist/docs/assets/chunks/{graph.V2GLaab4.js → graph.CNRB6ETL.js} +1 -1
- package/dist/docs/assets/chunks/{infoDiagram-42DDH7IO.7pZOkCCU.js → infoDiagram-42DDH7IO.CqhknMWi.js} +1 -1
- package/dist/docs/assets/chunks/{ishikawaDiagram-UXIWVN3A.DSMo3Qa3.js → ishikawaDiagram-UXIWVN3A.C6xpR2af.js} +1 -1
- package/dist/docs/assets/chunks/{journeyDiagram-VCZTEJTY.BNEgWN1S.js → journeyDiagram-VCZTEJTY.Cg5f7oB3.js} +1 -1
- package/dist/docs/assets/chunks/{kanban-definition-6JOO6SKY.B-4d4tC7.js → kanban-definition-6JOO6SKY.Cx9YTwlU.js} +1 -1
- package/dist/docs/assets/chunks/{layout.Dvbn9nSb.js → layout.ljS-wFtK.js} +1 -1
- package/dist/docs/assets/chunks/{linear.D2GM4p4b.js → linear.jSxNrsFC.js} +1 -1
- package/dist/docs/assets/chunks/{min.DVBtLK-B.js → min.Cum8AlQw.js} +1 -1
- package/dist/docs/assets/chunks/{mindmap-definition-QFDTVHPH.CnJRI55x.js → mindmap-definition-QFDTVHPH.BLiysLpe.js} +1 -1
- package/dist/docs/assets/chunks/{pieDiagram-DEJITSTG.CtoaTFlA.js → pieDiagram-DEJITSTG.BoIDyuKF.js} +1 -1
- package/dist/docs/assets/chunks/{quadrantDiagram-34T5L4WZ.DtT24_vN.js → quadrantDiagram-34T5L4WZ.DLkpDytR.js} +1 -1
- package/dist/docs/assets/chunks/{requirementDiagram-MS252O5E.CjSO4o8f.js → requirementDiagram-MS252O5E.DqTVqSu2.js} +1 -1
- package/dist/docs/assets/chunks/{sankeyDiagram-XADWPNL6.-wWiIVWa.js → sankeyDiagram-XADWPNL6.CG_6FF7j.js} +1 -1
- package/dist/docs/assets/chunks/{sequenceDiagram-FGHM5R23.DSk8s4gX.js → sequenceDiagram-FGHM5R23.BIp9602K.js} +1 -1
- package/dist/docs/assets/chunks/{stateDiagram-FHFEXIEX.BFKAsdkN.js → stateDiagram-FHFEXIEX.COSXsD9I.js} +1 -1
- package/dist/docs/assets/chunks/stateDiagram-v2-QKLJ7IA2.qrxrbFsX.js +1 -0
- package/dist/docs/assets/chunks/{theme.C0MctGaz.js → theme.CXJ7PNwy.js} +2 -2
- package/dist/docs/assets/chunks/{timeline-definition-GMOUNBTQ.TUNJbAFe.js → timeline-definition-GMOUNBTQ.CXdVqkLq.js} +1 -1
- package/dist/docs/assets/chunks/{vennDiagram-DHZGUBPP.C-RGNnk4.js → vennDiagram-DHZGUBPP.CZxGuc4r.js} +1 -1
- package/dist/docs/assets/chunks/{wardley-RL74JXVD.g5efOWmT.js → wardley-RL74JXVD.3oVgfqQk.js} +1 -1
- package/dist/docs/assets/chunks/{wardleyDiagram-NUSXRM2D.DH4zt73Y.js → wardleyDiagram-NUSXRM2D.6_irCgGJ.js} +1 -1
- package/dist/docs/assets/chunks/{xychartDiagram-5P7HB3ND.B-3k6bF6.js → xychartDiagram-5P7HB3ND.TRPe92m3.js} +1 -1
- package/dist/docs/assets/deployment.md.D2jQZuFx.js +32 -0
- package/dist/docs/assets/deployment.md.D2jQZuFx.lean.js +1 -0
- package/dist/docs/assets/evals.md.D3Y3Aixt.js +72 -0
- package/dist/docs/assets/evals.md.D3Y3Aixt.lean.js +1 -0
- package/dist/docs/assets/guides_agent-to-agent.md.CD4T5FIl.js +41 -0
- package/dist/docs/assets/guides_agent-to-agent.md.CD4T5FIl.lean.js +1 -0
- package/dist/docs/assets/guides_bitbucket.md.mpevW-VP.js +145 -0
- package/dist/docs/assets/guides_bitbucket.md.mpevW-VP.lean.js +1 -0
- package/dist/docs/assets/guides_cloud-agents.md.Cp1O3u-X.js +15 -0
- package/dist/docs/assets/guides_cloud-agents.md.Cp1O3u-X.lean.js +1 -0
- package/dist/docs/assets/guides_convert-automation.md.CqEyfP6Y.js +43 -0
- package/dist/docs/assets/guides_convert-automation.md.CqEyfP6Y.lean.js +1 -0
- package/dist/docs/assets/guides_github.md.BwpBp3ed.js +156 -0
- package/dist/docs/assets/guides_github.md.BwpBp3ed.lean.js +1 -0
- package/dist/docs/assets/guides_gitlab.md.DaEC3nMk.js +153 -0
- package/dist/docs/assets/guides_gitlab.md.DaEC3nMk.lean.js +1 -0
- package/dist/docs/assets/guides_grokbot-agents.md.CMhZNdEU.js +16 -0
- package/dist/docs/assets/guides_grokbot-agents.md.CMhZNdEU.lean.js +1 -0
- package/dist/docs/assets/guides_improve.md.Bnp4F99w.js +22 -0
- package/dist/docs/assets/guides_improve.md.Bnp4F99w.lean.js +1 -0
- package/dist/docs/assets/guides_jev.md.F5fAkkfN.js +189 -0
- package/dist/docs/assets/guides_jev.md.F5fAkkfN.lean.js +1 -0
- package/dist/docs/assets/guides_mcp-oauth.md.bSFakfCY.js +50 -0
- package/dist/docs/assets/guides_mcp-oauth.md.bSFakfCY.lean.js +1 -0
- package/dist/docs/assets/guides_opentelemetry.md.BKDxQmmd.js +35 -0
- package/dist/docs/assets/guides_opentelemetry.md.BKDxQmmd.lean.js +1 -0
- package/dist/docs/assets/guides_slack.md.Bo96y42E.js +70 -0
- package/dist/docs/assets/guides_slack.md.Bo96y42E.lean.js +1 -0
- package/dist/docs/assets/guides_webhooks.md.1A72_VEE.js +92 -0
- package/dist/docs/assets/guides_webhooks.md.1A72_VEE.lean.js +1 -0
- package/dist/docs/assets/hillclimbing.md.D4E1o5Sa.js +7 -0
- package/dist/docs/assets/hillclimbing.md.D4E1o5Sa.lean.js +1 -0
- package/dist/docs/assets/{index.md.BFVyY2KT.js → index.md.C-t81M5J.js} +2 -2
- package/dist/docs/assets/{index.md.BFVyY2KT.lean.js → index.md.C-t81M5J.lean.js} +1 -1
- package/dist/docs/assets/{quickstart.md.D3MjSZN-.js → quickstart.md.DAvVhuuU.js} +1 -1
- package/dist/docs/assets/{quickstart.md.D3MjSZN-.lean.js → quickstart.md.DAvVhuuU.lean.js} +1 -1
- package/dist/docs/assets/{reference_agent-config.md.CfVA-LZJ.js → reference_agent-config.md.DGPyw7ms.js} +1 -1
- package/dist/docs/assets/{reference_agent-config.md.CfVA-LZJ.lean.js → reference_agent-config.md.DGPyw7ms.lean.js} +1 -1
- package/dist/docs/assets/{reference_artifacts.md.Vf7qyIZ-.js → reference_artifacts.md.Bu_4HmsD.js} +1 -1
- package/dist/docs/assets/{reference_artifacts.md.Vf7qyIZ-.lean.js → reference_artifacts.md.Bu_4HmsD.lean.js} +1 -1
- package/dist/docs/assets/{reference_channels.md.icqLKcTc.js → reference_channels.md.nFWbzAic.js} +1 -1
- package/dist/docs/assets/{reference_channels.md.icqLKcTc.lean.js → reference_channels.md.nFWbzAic.lean.js} +1 -1
- package/dist/docs/assets/{reference_cli.md.B2dBL6L8.js → reference_cli.md.DLWDz9ij.js} +3 -1
- package/dist/docs/assets/{reference_cli.md.B2dBL6L8.lean.js → reference_cli.md.DLWDz9ij.lean.js} +1 -1
- package/dist/docs/assets/{reference_connections.md.Cb3U_c8n.js → reference_connections.md.Je9dMsdd.js} +2 -2
- package/dist/docs/assets/{reference_connections.md.Cb3U_c8n.lean.js → reference_connections.md.Je9dMsdd.lean.js} +1 -1
- package/dist/docs/assets/reference_evals.md.DNJzM_yf.js +57 -0
- package/dist/docs/assets/reference_evals.md.DNJzM_yf.lean.js +1 -0
- package/dist/docs/assets/reference_extensions.md.Cv5aLCz_.js +58 -0
- package/dist/docs/assets/reference_extensions.md.Cv5aLCz_.lean.js +1 -0
- package/dist/docs/assets/{reference_hooks.md.CZuynAxj.js → reference_hooks.md.B7uzNENk.js} +2 -2
- package/dist/docs/assets/{reference_hooks.md.CZuynAxj.lean.js → reference_hooks.md.B7uzNENk.lean.js} +1 -1
- package/dist/docs/assets/{reference_http-api.md.DTKcYE6L.js → reference_http-api.md.CduHavZ2.js} +1 -1
- package/dist/docs/assets/{reference_http-api.md.DTKcYE6L.lean.js → reference_http-api.md.CduHavZ2.lean.js} +1 -1
- package/dist/docs/assets/{reference_instructions.md.D7gkckK-.js → reference_instructions.md.CU1My5My.js} +1 -1
- package/dist/docs/assets/{reference_instructions.md.D7gkckK-.lean.js → reference_instructions.md.CU1My5My.lean.js} +1 -1
- package/dist/docs/assets/{reference_playground.md.D2YExv5K.js → reference_playground.md.Ch2d0Iqi.js} +1 -1
- package/dist/docs/assets/{reference_playground.md.D2YExv5K.lean.js → reference_playground.md.Ch2d0Iqi.lean.js} +1 -1
- package/dist/docs/assets/{reference_project-layout.md.DPxbUJyt.js → reference_project-layout.md.BGhgpy9V.js} +1 -1
- package/dist/docs/assets/{reference_project-layout.md.DPxbUJyt.lean.js → reference_project-layout.md.BGhgpy9V.lean.js} +1 -1
- package/dist/docs/assets/{reference_prompt.md.BQ5uAv1F.js → reference_prompt.md.Ccp0R53H.js} +1 -1
- package/dist/docs/assets/{reference_prompt.md.BQ5uAv1F.lean.js → reference_prompt.md.Ccp0R53H.lean.js} +1 -1
- package/dist/docs/assets/{reference_schedules.md.BasfZWO-.js → reference_schedules.md.B2Nm6FaD.js} +1 -1
- package/dist/docs/assets/{reference_schedules.md.BasfZWO-.lean.js → reference_schedules.md.B2Nm6FaD.lean.js} +1 -1
- package/dist/docs/assets/{reference_sessions.md.YKvIsWAx.js → reference_sessions.md.1_6Vyv7x.js} +1 -1
- package/dist/docs/assets/{reference_sessions.md.YKvIsWAx.lean.js → reference_sessions.md.1_6Vyv7x.lean.js} +1 -1
- package/dist/docs/assets/{reference_skills.md.rNgpsGd0.js → reference_skills.md.DjQkRefx.js} +1 -1
- package/dist/docs/assets/{reference_skills.md.rNgpsGd0.lean.js → reference_skills.md.DjQkRefx.lean.js} +1 -1
- package/dist/docs/assets/{reference_subagents.md.e5qitjJt.js → reference_subagents.md.Dl16gcBj.js} +2 -2
- package/dist/docs/assets/{reference_subagents.md.e5qitjJt.lean.js → reference_subagents.md.Dl16gcBj.lean.js} +1 -1
- package/dist/docs/assets/{reference_tools.md.BdCO2aHZ.js → reference_tools.md.B1dH1lpa.js} +3 -3
- package/dist/docs/assets/{reference_tools.md.BdCO2aHZ.lean.js → reference_tools.md.B1dH1lpa.lean.js} +1 -1
- package/dist/docs/assets/{templates_agentic-owners.md.Da_AGDlH.js → templates_agentic-owners.md.9M575F5C.js} +1 -1
- package/dist/docs/assets/{templates_agentic-owners.md.Da_AGDlH.lean.js → templates_agentic-owners.md.9M575F5C.lean.js} +1 -1
- package/dist/docs/assets/{templates_pr-autofixer.md.DqxocIGh.js → templates_pr-autofixer.md.ws0DDXDy.js} +1 -1
- package/dist/docs/assets/{templates_pr-autofixer.md.DqxocIGh.lean.js → templates_pr-autofixer.md.ws0DDXDy.lean.js} +1 -1
- package/dist/docs/assets/{templates_security-reviewer.md.Bhnvd8VE.js → templates_security-reviewer.md.KEFYXzfK.js} +1 -1
- package/dist/docs/assets/{templates_security-reviewer.md.Bhnvd8VE.lean.js → templates_security-reviewer.md.KEFYXzfK.lean.js} +1 -1
- package/dist/docs/assets/templates_thermo-quality-review.md.VNJ_mohX.js +3 -0
- package/dist/docs/assets/templates_thermo-quality-review.md.VNJ_mohX.lean.js +1 -0
- package/dist/docs/assets/templates_thermo-review.md.Hi3zWOkP.js +3 -0
- package/dist/docs/assets/templates_thermo-review.md.Hi3zWOkP.lean.js +1 -0
- package/dist/docs/assets/{templates_triage.md.BdBWO9Ic.js → templates_triage.md.CVGe_FG6.js} +1 -1
- package/dist/docs/assets/{templates_triage.md.BdBWO9Ic.lean.js → templates_triage.md.CVGe_FG6.lean.js} +1 -1
- package/dist/docs/assets/troubleshooting.md.mnfFG2Em.js +1 -0
- package/dist/docs/assets/troubleshooting.md.mnfFG2Em.lean.js +1 -0
- package/dist/docs/building-with-agents.html +44 -48
- package/dist/docs/building-with-agents.md +94 -82
- package/dist/docs/deployment.html +59 -78
- package/dist/docs/deployment.md +117 -363
- package/dist/docs/evals.html +85 -224
- package/dist/docs/evals.md +149 -673
- package/dist/docs/guides/agent-to-agent.html +74 -47
- package/dist/docs/guides/agent-to-agent.md +128 -46
- package/dist/docs/guides/bitbucket.html +179 -44
- package/dist/docs/guides/bitbucket.md +249 -48
- package/dist/docs/guides/cloud-agents.html +46 -40
- package/dist/docs/guides/cloud-agents.md +119 -66
- package/dist/docs/guides/convert-automation.html +80 -49
- package/dist/docs/guides/convert-automation.md +156 -147
- package/dist/docs/guides/github.html +184 -92
- package/dist/docs/guides/github.md +260 -245
- package/dist/docs/guides/gitlab.html +184 -45
- package/dist/docs/guides/gitlab.md +249 -50
- package/dist/docs/guides/grokbot-agents.html +48 -41
- package/dist/docs/guides/grokbot-agents.md +99 -53
- package/dist/docs/guides/improve.html +48 -40
- package/dist/docs/guides/improve.md +111 -58
- package/dist/docs/guides/jev.html +248 -0
- package/dist/docs/guides/jev.md +348 -0
- package/dist/docs/guides/mcp-oauth.html +74 -52
- package/dist/docs/guides/mcp-oauth.md +111 -121
- package/dist/docs/guides/opentelemetry.html +63 -54
- package/dist/docs/guides/opentelemetry.md +96 -165
- package/dist/docs/guides/slack.html +83 -59
- package/dist/docs/guides/slack.md +157 -227
- package/dist/docs/guides/webhooks.html +97 -230
- package/dist/docs/guides/webhooks.md +154 -385
- package/dist/docs/hashmap.json +1 -1
- package/dist/docs/hillclimbing.html +44 -38
- package/dist/docs/hillclimbing.md +101 -55
- package/dist/docs/index.html +38 -38
- package/dist/docs/index.md +2 -0
- package/dist/docs/llms-full.txt +3490 -3146
- package/dist/docs/llms.txt +22 -18
- package/dist/docs/quickstart.html +37 -37
- package/dist/docs/reference/agent-config.html +37 -37
- package/dist/docs/reference/artifacts.html +38 -38
- package/dist/docs/reference/channels.html +37 -37
- package/dist/docs/reference/cli.html +39 -37
- package/dist/docs/reference/cli.md +2 -0
- package/dist/docs/reference/connections.html +38 -38
- package/dist/docs/reference/connections.md +1 -1
- package/dist/docs/reference/evals.html +116 -0
- package/dist/docs/reference/evals.md +293 -0
- package/dist/docs/reference/extensions.html +80 -84
- package/dist/docs/reference/extensions.md +131 -202
- package/dist/docs/reference/hooks.html +39 -39
- package/dist/docs/reference/hooks.md +34 -34
- package/dist/docs/reference/http-api.html +37 -37
- package/dist/docs/reference/instructions.html +37 -37
- package/dist/docs/reference/playground.html +37 -37
- package/dist/docs/reference/project-layout.html +37 -37
- package/dist/docs/reference/prompt.html +37 -37
- package/dist/docs/reference/schedules.html +37 -37
- package/dist/docs/reference/sessions.html +37 -37
- package/dist/docs/reference/skills.html +37 -37
- package/dist/docs/reference/subagents.html +38 -38
- package/dist/docs/reference/subagents.md +1 -1
- package/dist/docs/reference/tools.html +38 -38
- package/dist/docs/reference/tools.md +4 -0
- package/dist/docs/templates/agentic-owners.html +38 -38
- package/dist/docs/templates/pr-autofixer.html +37 -37
- package/dist/docs/templates/security-reviewer.html +38 -38
- package/dist/docs/templates/thermo-quality-review.html +62 -0
- package/dist/docs/templates/thermo-quality-review.md +75 -0
- package/dist/docs/templates/thermo-review.html +62 -0
- package/dist/docs/templates/thermo-review.md +74 -0
- package/dist/docs/templates/triage.html +37 -37
- package/dist/docs/troubleshooting.html +38 -38
- package/dist/docs/troubleshooting.md +97 -66
- package/dist/extensions/cursor-cloud-agents/skills/handoff.md +3 -7
- package/dist/extensions/improve/extension.d.ts +3 -1
- package/dist/extensions/improve/extension.d.ts.map +1 -1
- package/dist/extensions/improve/extension.js +4 -2
- package/dist/extensions/improve/skills/yourself.js +1 -1
- package/dist/extensions/jev/extension.d.ts +43 -0
- package/dist/extensions/jev/extension.d.ts.map +1 -0
- package/dist/extensions/jev/extension.js +47 -0
- package/dist/extensions/jev/lib/evaluate.d.ts +101 -0
- package/dist/extensions/jev/lib/evaluate.d.ts.map +1 -0
- package/dist/extensions/jev/lib/evaluate.js +167 -0
- package/dist/extensions/jev/skills/gated-write.md +25 -0
- package/dist/extensions/jev/skills/questions.md +33 -0
- package/dist/extensions/jev/tools/evaluate.d.ts +4 -0
- package/dist/extensions/jev/tools/evaluate.d.ts.map +1 -0
- package/dist/extensions/jev/tools/evaluate.js +88 -0
- package/dist/extensions.d.ts +1 -1
- package/dist/extensions.d.ts.map +1 -1
- package/dist/extensions.js +2 -0
- package/dist/filesystem.d.ts +46 -2
- package/dist/filesystem.d.ts.map +1 -1
- package/dist/filesystem.js +149 -102
- package/dist/index.d.ts +2 -2
- package/dist/index.d.ts.map +1 -1
- package/dist/index.js +1 -1
- package/dist/internal/advertise-tools.d.ts.map +1 -1
- package/dist/internal/advertise-tools.js +6 -0
- package/dist/internal/cli-ax.d.ts +6 -0
- package/dist/internal/cli-ax.d.ts.map +1 -1
- package/dist/internal/cli-ax.js +72 -14
- package/dist/internal/continuation-identity.d.ts.map +1 -1
- package/dist/internal/continuation-identity.js +1 -0
- package/dist/internal/cursor-agent-template.d.ts +1 -1
- package/dist/internal/cursor-agent-template.d.ts.map +1 -1
- package/dist/internal/cursor-agent-template.js +2 -0
- package/dist/internal/discovery/connections.d.ts.map +1 -1
- package/dist/internal/discovery/connections.js +18 -0
- package/dist/internal/discovery/extensions.d.ts.map +1 -1
- package/dist/internal/discovery/extensions.js +8 -4
- package/dist/internal/discovery/info.d.ts.map +1 -1
- package/dist/internal/discovery/info.js +1 -0
- package/dist/internal/filesystem/tools.d.ts.map +1 -1
- package/dist/internal/filesystem/tools.js +2 -2
- package/dist/internal/filesystem/walk.d.ts +7 -4
- package/dist/internal/filesystem/walk.d.ts.map +1 -1
- package/dist/internal/filesystem/walk.js +34 -11
- package/dist/internal/hosted-admission-adapter.d.ts +3 -0
- package/dist/internal/hosted-admission-adapter.d.ts.map +1 -1
- package/dist/internal/hosted-delivery-protocol.d.ts +20 -0
- package/dist/internal/hosted-delivery-protocol.d.ts.map +1 -1
- package/dist/internal/hosted-delivery-protocol.js +51 -1
- package/dist/internal/hosted-delivery.d.ts.map +1 -1
- package/dist/internal/hosted-delivery.js +23 -25
- package/dist/internal/hosted-execution-diag.d.ts +12 -4
- package/dist/internal/hosted-execution-diag.d.ts.map +1 -1
- package/dist/internal/hosted-execution-diag.js +26 -4
- package/dist/internal/hosted-execution-flush.d.ts +1 -0
- package/dist/internal/hosted-execution-flush.d.ts.map +1 -1
- package/dist/internal/hosted-execution-flush.js +4 -2
- package/dist/internal/init-project.d.ts.map +1 -1
- package/dist/internal/init-project.js +4 -0
- package/dist/internal/server.d.ts.map +1 -1
- package/dist/internal/server.js +111 -48
- package/dist/internal/session-engine.d.ts +4 -1
- package/dist/internal/session-engine.d.ts.map +1 -1
- package/dist/internal/session-engine.js +39 -9
- package/dist/internal/skill-catalog.d.ts +28 -0
- package/dist/internal/skill-catalog.d.ts.map +1 -0
- package/dist/internal/skill-catalog.js +44 -0
- package/dist/playground/assets/index-DSMAewbx.css +1 -0
- package/dist/playground/assets/index-De_lpFxE.js +67 -0
- package/dist/playground/index.html +2 -2
- package/dist/types.d.ts +32 -3
- package/dist/types.d.ts.map +1 -1
- package/docs/README.md +2 -0
- package/docs/building-with-agents.md +96 -84
- package/docs/deployment.md +118 -364
- package/docs/evals.md +149 -673
- package/docs/guides/agent-to-agent.md +130 -48
- package/docs/guides/bitbucket.md +250 -49
- package/docs/guides/cloud-agents.md +119 -67
- package/docs/guides/convert-automation.md +157 -148
- package/docs/guides/github.md +261 -246
- package/docs/guides/gitlab.md +250 -51
- package/docs/guides/grokbot-agents.md +100 -55
- package/docs/guides/improve.md +112 -59
- package/docs/guides/jev.md +353 -0
- package/docs/guides/mcp-oauth.md +112 -122
- package/docs/guides/opentelemetry.md +97 -166
- package/docs/guides/slack.md +158 -228
- package/docs/guides/webhooks.md +155 -386
- package/docs/hillclimbing.md +102 -56
- package/docs/reference/cli.md +2 -0
- package/docs/reference/connections.md +1 -1
- package/docs/reference/evals.md +298 -0
- package/docs/reference/extensions.md +132 -203
- package/docs/reference/hooks.md +34 -34
- package/docs/reference/subagents.md +1 -1
- package/docs/reference/tools.md +4 -0
- package/docs/templates/thermo-quality-review.md +80 -0
- package/docs/templates/thermo-review.md +79 -0
- package/docs/troubleshooting.md +98 -67
- package/package.json +8 -1
- package/skills/github/SKILL.md +21 -13
- package/src/bin/agent-serve.ts +24 -3
- package/src/channels/checks.ts +8 -0
- package/src/channels/github/github-channel.ts +29 -19
- package/src/channels/origin/checks.ts +3 -1
- package/src/channels/slack/agentic-delivery.ts +2 -0
- package/src/channels/slack/dispatch.ts +131 -4
- package/src/channels/slack/placeholder.ts +51 -0
- package/src/channels/slack/slack-channel.ts +9 -2
- package/src/extensions/cursor-cloud-agents/skills/handoff.md +3 -7
- package/src/extensions/improve/extension.ts +4 -2
- package/src/extensions/improve/skills/yourself.ts +1 -1
- package/src/extensions/jev/extension.ts +95 -0
- package/src/extensions/jev/lib/evaluate.ts +289 -0
- package/src/extensions/jev/skills/gated-write.md +25 -0
- package/src/extensions/jev/skills/questions.md +33 -0
- package/src/extensions/jev/tools/evaluate.ts +90 -0
- package/src/extensions.ts +2 -0
- package/src/filesystem.ts +168 -65
- package/src/index.ts +3 -0
- package/src/internal/advertise-tools.ts +6 -0
- package/src/internal/cli-ax.ts +88 -15
- package/src/internal/continuation-identity.ts +1 -0
- package/src/internal/cursor-agent-template.ts +2 -0
- package/src/internal/discovery/connections.ts +21 -0
- package/src/internal/discovery/extensions.ts +12 -4
- package/src/internal/discovery/info.ts +1 -0
- package/src/internal/filesystem/tools.ts +2 -0
- package/src/internal/filesystem/walk.ts +60 -15
- package/src/internal/hosted-admission-adapter.ts +3 -0
- package/src/internal/hosted-delivery-protocol.ts +82 -1
- package/src/internal/hosted-delivery.ts +23 -0
- package/src/internal/hosted-execution-diag.ts +33 -4
- package/src/internal/hosted-execution-flush.ts +4 -0
- package/src/internal/init-project.ts +4 -0
- package/src/internal/server.ts +130 -53
- package/src/internal/session-engine.ts +46 -8
- package/src/internal/skill-catalog.ts +69 -0
- package/src/types.ts +33 -3
- package/templates/thermo-quality-review/README.md +35 -0
- package/templates/thermo-quality-review/agent/agent.ts +8 -0
- package/templates/thermo-quality-review/agent/channels/github.ts +42 -0
- package/templates/thermo-quality-review/agent/instructions.md +43 -0
- package/templates/thermo-quality-review/agent/tools/post_findings.ts +70 -0
- package/templates/thermo-quality-review/evals/evals.config.ts +5 -0
- package/templates/thermo-quality-review/evals/review.eval.ts +62 -0
- package/templates/thermo-quality-review/package.json +18 -0
- package/templates/thermo-quality-review/tsconfig.json +12 -0
- package/templates/thermo-review/README.md +35 -0
- package/templates/thermo-review/agent/agent.ts +8 -0
- package/templates/thermo-review/agent/channels/github.ts +42 -0
- package/templates/thermo-review/agent/instructions.md +40 -0
- package/templates/thermo-review/agent/tools/post_findings.ts +70 -0
- package/templates/thermo-review/evals/evals.config.ts +5 -0
- package/templates/thermo-review/evals/review.eval.ts +53 -0
- package/templates/thermo-review/package.json +18 -0
- package/templates/thermo-review/tsconfig.json +12 -0
- package/dist/docs/assets/building-with-agents.md.CEGVXkmO.js +0 -13
- package/dist/docs/assets/building-with-agents.md.CEGVXkmO.lean.js +0 -1
- package/dist/docs/assets/chunks/@localSearchIndexroot.DtXk1hy-.js +0 -1
- package/dist/docs/assets/chunks/channel.Bkv1N-gK.js +0 -1
- package/dist/docs/assets/chunks/classDiagram-6PBFFD2Q.CZDco1o8.js +0 -1
- package/dist/docs/assets/chunks/classDiagram-v2-HSJHXN6E.CZDco1o8.js +0 -1
- package/dist/docs/assets/chunks/clone.YSt_40_s.js +0 -1
- package/dist/docs/assets/chunks/framework.dypDpWZ3.js +0 -19
- package/dist/docs/assets/chunks/stateDiagram-v2-QKLJ7IA2.C4CLm481.js +0 -1
- package/dist/docs/assets/deployment.md.Dm4Qo3hp.js +0 -51
- package/dist/docs/assets/deployment.md.Dm4Qo3hp.lean.js +0 -1
- package/dist/docs/assets/evals.md.BLDRt5LH.js +0 -211
- package/dist/docs/assets/evals.md.BLDRt5LH.lean.js +0 -1
- package/dist/docs/assets/guides_agent-to-agent.md.BI0xclmy.js +0 -14
- package/dist/docs/assets/guides_agent-to-agent.md.BI0xclmy.lean.js +0 -1
- package/dist/docs/assets/guides_bitbucket.md.CTpCl__f.js +0 -10
- package/dist/docs/assets/guides_bitbucket.md.CTpCl__f.lean.js +0 -1
- package/dist/docs/assets/guides_cloud-agents.md.lSE_l7lH.js +0 -9
- package/dist/docs/assets/guides_cloud-agents.md.lSE_l7lH.lean.js +0 -1
- package/dist/docs/assets/guides_convert-automation.md.Ck6Cr68A.js +0 -12
- package/dist/docs/assets/guides_convert-automation.md.Ck6Cr68A.lean.js +0 -1
- package/dist/docs/assets/guides_github.md.D6ER29dG.js +0 -64
- package/dist/docs/assets/guides_github.md.D6ER29dG.lean.js +0 -1
- package/dist/docs/assets/guides_gitlab.md.P-TjBnS5.js +0 -14
- package/dist/docs/assets/guides_gitlab.md.P-TjBnS5.lean.js +0 -1
- package/dist/docs/assets/guides_grokbot-agents.md.WBZIOvkz.js +0 -9
- package/dist/docs/assets/guides_grokbot-agents.md.WBZIOvkz.lean.js +0 -1
- package/dist/docs/assets/guides_improve.md.BKaDuKKK.js +0 -14
- package/dist/docs/assets/guides_improve.md.BKaDuKKK.lean.js +0 -1
- package/dist/docs/assets/guides_mcp-oauth.md.DMNMpXtO.js +0 -28
- package/dist/docs/assets/guides_mcp-oauth.md.DMNMpXtO.lean.js +0 -1
- package/dist/docs/assets/guides_opentelemetry.md._CRfDyzH.js +0 -26
- package/dist/docs/assets/guides_opentelemetry.md._CRfDyzH.lean.js +0 -1
- package/dist/docs/assets/guides_slack.md.DdT8rmsj.js +0 -46
- package/dist/docs/assets/guides_slack.md.DdT8rmsj.lean.js +0 -1
- package/dist/docs/assets/guides_webhooks.md.aQW10HRe.js +0 -225
- package/dist/docs/assets/guides_webhooks.md.aQW10HRe.lean.js +0 -1
- package/dist/docs/assets/hillclimbing.md.Dq4kkVIL.js +0 -1
- package/dist/docs/assets/hillclimbing.md.Dq4kkVIL.lean.js +0 -1
- package/dist/docs/assets/reference_extensions.md.wlFD3cUR.js +0 -62
- package/dist/docs/assets/reference_extensions.md.wlFD3cUR.lean.js +0 -1
- package/dist/docs/assets/troubleshooting.md.BcgNoYtJ.js +0 -1
- package/dist/docs/assets/troubleshooting.md.BcgNoYtJ.lean.js +0 -1
- package/dist/playground/assets/index-BLlKgZtI.css +0 -1
- package/dist/playground/assets/index-CDS5p9sR.js +0 -67
|
@@ -4,57 +4,63 @@
|
|
|
4
4
|
<meta charset="utf-8">
|
|
5
5
|
<meta name="viewport" content="width=device-width,initial-scale=1">
|
|
6
6
|
<title>Hillclimbing | Agent SDK</title>
|
|
7
|
-
<meta name="description" content="Improve
|
|
7
|
+
<meta name="description" content="Improve one agent behavior at a time by measuring fixed inputs, changing one lever, and locking each win with an eval.">
|
|
8
8
|
<meta name="generator" content="VitePress v1.6.4">
|
|
9
9
|
<link rel="preload stylesheet" href="/docs/assets/style.Cfo2vbik.css" as="style">
|
|
10
10
|
<link rel="preload stylesheet" href="/docs/vp-icons.css" as="style">
|
|
11
11
|
|
|
12
|
-
<script type="module" src="/docs/assets/app.
|
|
12
|
+
<script type="module" src="/docs/assets/app.CAeK13eM.js"></script>
|
|
13
13
|
<link rel="preload" href="/docs/assets/inter-roman-latin.Di8DUHzh.woff2" as="font" type="font/woff2" crossorigin="">
|
|
14
|
-
<link rel="modulepreload" href="/docs/assets/chunks/framework.
|
|
15
|
-
<link rel="modulepreload" href="/docs/assets/chunks/theme.
|
|
14
|
+
<link rel="modulepreload" href="/docs/assets/chunks/framework.BNw1pucY.js">
|
|
15
|
+
<link rel="modulepreload" href="/docs/assets/chunks/theme.CXJ7PNwy.js">
|
|
16
16
|
<link rel="modulepreload" href="/docs/assets/chunks/katex.DkKDou_j.js">
|
|
17
|
-
<link rel="modulepreload" href="/docs/assets/chunks/dagre-KV5264BT.
|
|
18
|
-
<link rel="modulepreload" href="/docs/assets/chunks/cose-bilkent-S5V4N54A.
|
|
19
|
-
<link rel="modulepreload" href="/docs/assets/chunks/c4Diagram-AHTNJAMY.
|
|
20
|
-
<link rel="modulepreload" href="/docs/assets/chunks/flowDiagram-DWJPFMVM.
|
|
21
|
-
<link rel="modulepreload" href="/docs/assets/chunks/erDiagram-SMLLAGMA
|
|
22
|
-
<link rel="modulepreload" href="/docs/assets/chunks/gitGraphDiagram-UUTBAWPF.
|
|
23
|
-
<link rel="modulepreload" href="/docs/assets/chunks/ganttDiagram-T4ZO3ILL.
|
|
24
|
-
<link rel="modulepreload" href="/docs/assets/chunks/infoDiagram-42DDH7IO.
|
|
25
|
-
<link rel="modulepreload" href="/docs/assets/chunks/pieDiagram-DEJITSTG.
|
|
26
|
-
<link rel="modulepreload" href="/docs/assets/chunks/quadrantDiagram-34T5L4WZ.
|
|
27
|
-
<link rel="modulepreload" href="/docs/assets/chunks/xychartDiagram-5P7HB3ND.
|
|
28
|
-
<link rel="modulepreload" href="/docs/assets/chunks/requirementDiagram-MS252O5E.
|
|
29
|
-
<link rel="modulepreload" href="/docs/assets/chunks/sequenceDiagram-FGHM5R23.
|
|
30
|
-
<link rel="modulepreload" href="/docs/assets/chunks/classDiagram-6PBFFD2Q.
|
|
31
|
-
<link rel="modulepreload" href="/docs/assets/chunks/classDiagram-v2-HSJHXN6E.
|
|
32
|
-
<link rel="modulepreload" href="/docs/assets/chunks/stateDiagram-FHFEXIEX.
|
|
33
|
-
<link rel="modulepreload" href="/docs/assets/chunks/stateDiagram-v2-QKLJ7IA2.
|
|
34
|
-
<link rel="modulepreload" href="/docs/assets/chunks/journeyDiagram-VCZTEJTY.
|
|
35
|
-
<link rel="modulepreload" href="/docs/assets/chunks/timeline-definition-GMOUNBTQ.
|
|
36
|
-
<link rel="modulepreload" href="/docs/assets/chunks/mindmap-definition-QFDTVHPH.
|
|
37
|
-
<link rel="modulepreload" href="/docs/assets/chunks/kanban-definition-6JOO6SKY.
|
|
38
|
-
<link rel="modulepreload" href="/docs/assets/chunks/sankeyDiagram-XADWPNL6
|
|
39
|
-
<link rel="modulepreload" href="/docs/assets/chunks/diagram-TYMM5635.
|
|
40
|
-
<link rel="modulepreload" href="/docs/assets/chunks/diagram-MMDJMWI5.
|
|
41
|
-
<link rel="modulepreload" href="/docs/assets/chunks/blockDiagram-DXYQGD6D.
|
|
42
|
-
<link rel="modulepreload" href="/docs/assets/chunks/diagram-5BDNPKRD.
|
|
43
|
-
<link rel="modulepreload" href="/docs/assets/chunks/architectureDiagram-Q4EWVU46.
|
|
44
|
-
<link rel="modulepreload" href="/docs/assets/chunks/ishikawaDiagram-UXIWVN3A.
|
|
45
|
-
<link rel="modulepreload" href="/docs/assets/chunks/vennDiagram-DHZGUBPP.
|
|
46
|
-
<link rel="modulepreload" href="/docs/assets/chunks/diagram-G4DWMVQ6.
|
|
47
|
-
<link rel="modulepreload" href="/docs/assets/chunks/wardleyDiagram-NUSXRM2D.
|
|
17
|
+
<link rel="modulepreload" href="/docs/assets/chunks/dagre-KV5264BT.C9PZQK-S.js">
|
|
18
|
+
<link rel="modulepreload" href="/docs/assets/chunks/cose-bilkent-S5V4N54A.DVEa6fZp.js">
|
|
19
|
+
<link rel="modulepreload" href="/docs/assets/chunks/c4Diagram-AHTNJAMY.BRV8GPJJ.js">
|
|
20
|
+
<link rel="modulepreload" href="/docs/assets/chunks/flowDiagram-DWJPFMVM.ByWJXeYK.js">
|
|
21
|
+
<link rel="modulepreload" href="/docs/assets/chunks/erDiagram-SMLLAGMA.U2pR_OA7.js">
|
|
22
|
+
<link rel="modulepreload" href="/docs/assets/chunks/gitGraphDiagram-UUTBAWPF.Bpn01P7X.js">
|
|
23
|
+
<link rel="modulepreload" href="/docs/assets/chunks/ganttDiagram-T4ZO3ILL.OquF0Rtg.js">
|
|
24
|
+
<link rel="modulepreload" href="/docs/assets/chunks/infoDiagram-42DDH7IO.CqhknMWi.js">
|
|
25
|
+
<link rel="modulepreload" href="/docs/assets/chunks/pieDiagram-DEJITSTG.BoIDyuKF.js">
|
|
26
|
+
<link rel="modulepreload" href="/docs/assets/chunks/quadrantDiagram-34T5L4WZ.DLkpDytR.js">
|
|
27
|
+
<link rel="modulepreload" href="/docs/assets/chunks/xychartDiagram-5P7HB3ND.TRPe92m3.js">
|
|
28
|
+
<link rel="modulepreload" href="/docs/assets/chunks/requirementDiagram-MS252O5E.DqTVqSu2.js">
|
|
29
|
+
<link rel="modulepreload" href="/docs/assets/chunks/sequenceDiagram-FGHM5R23.BIp9602K.js">
|
|
30
|
+
<link rel="modulepreload" href="/docs/assets/chunks/classDiagram-6PBFFD2Q.Degh8l90.js">
|
|
31
|
+
<link rel="modulepreload" href="/docs/assets/chunks/classDiagram-v2-HSJHXN6E.Degh8l90.js">
|
|
32
|
+
<link rel="modulepreload" href="/docs/assets/chunks/stateDiagram-FHFEXIEX.COSXsD9I.js">
|
|
33
|
+
<link rel="modulepreload" href="/docs/assets/chunks/stateDiagram-v2-QKLJ7IA2.qrxrbFsX.js">
|
|
34
|
+
<link rel="modulepreload" href="/docs/assets/chunks/journeyDiagram-VCZTEJTY.Cg5f7oB3.js">
|
|
35
|
+
<link rel="modulepreload" href="/docs/assets/chunks/timeline-definition-GMOUNBTQ.CXdVqkLq.js">
|
|
36
|
+
<link rel="modulepreload" href="/docs/assets/chunks/mindmap-definition-QFDTVHPH.BLiysLpe.js">
|
|
37
|
+
<link rel="modulepreload" href="/docs/assets/chunks/kanban-definition-6JOO6SKY.Cx9YTwlU.js">
|
|
38
|
+
<link rel="modulepreload" href="/docs/assets/chunks/sankeyDiagram-XADWPNL6.CG_6FF7j.js">
|
|
39
|
+
<link rel="modulepreload" href="/docs/assets/chunks/diagram-TYMM5635.DW326M4K.js">
|
|
40
|
+
<link rel="modulepreload" href="/docs/assets/chunks/diagram-MMDJMWI5.BinJ5kWb.js">
|
|
41
|
+
<link rel="modulepreload" href="/docs/assets/chunks/blockDiagram-DXYQGD6D.JYq6w91N.js">
|
|
42
|
+
<link rel="modulepreload" href="/docs/assets/chunks/diagram-5BDNPKRD.DoN0uv3Y.js">
|
|
43
|
+
<link rel="modulepreload" href="/docs/assets/chunks/architectureDiagram-Q4EWVU46.CCXB8Uj5.js">
|
|
44
|
+
<link rel="modulepreload" href="/docs/assets/chunks/ishikawaDiagram-UXIWVN3A.C6xpR2af.js">
|
|
45
|
+
<link rel="modulepreload" href="/docs/assets/chunks/vennDiagram-DHZGUBPP.CZxGuc4r.js">
|
|
46
|
+
<link rel="modulepreload" href="/docs/assets/chunks/diagram-G4DWMVQ6.Czv3duqx.js">
|
|
47
|
+
<link rel="modulepreload" href="/docs/assets/chunks/wardleyDiagram-NUSXRM2D.6_irCgGJ.js">
|
|
48
48
|
<link rel="modulepreload" href="/docs/assets/chunks/virtual_mermaid-config.DDnGl6nM.js">
|
|
49
|
-
<link rel="modulepreload" href="/docs/assets/hillclimbing.md.
|
|
49
|
+
<link rel="modulepreload" href="/docs/assets/hillclimbing.md.D4E1o5Sa.lean.js">
|
|
50
50
|
<script id="check-dark-mode">(()=>{const e=localStorage.getItem("vitepress-theme-appearance")||"auto",a=window.matchMedia("(prefers-color-scheme: dark)").matches;(!e||e==="auto"?a:e==="dark")&&document.documentElement.classList.add("dark")})();</script>
|
|
51
51
|
<script id="check-mac-os">document.documentElement.classList.toggle("mac",/Mac|iPhone|iPod|iPad/i.test(navigator.platform));</script>
|
|
52
52
|
<link rel="alternate" type="text/plain" href="/docs/llms.txt">
|
|
53
53
|
<link rel="alternate" type="text/markdown" href="/docs/hillclimbing.md">
|
|
54
54
|
</head>
|
|
55
55
|
<body>
|
|
56
|
-
<div id="app"><div class="Layout" data-v-76086e96><!--[--><!--]--><!--[--><span tabindex="-1" data-v-97fb5898></span><a href="#VPContent" class="VPSkipLink visually-hidden" data-v-97fb5898>Skip to content</a><!--]--><!----><header class="VPNav" data-v-76086e96 data-v-781b1279><div class="VPNavBar" data-v-781b1279 data-v-b33015df><div class="wrapper" data-v-b33015df><div class="container" data-v-b33015df><div class="title" data-v-b33015df><div class="VPNavBarTitle has-sidebar" data-v-b33015df data-v-a0c466dc><a class="title" href="/docs/" data-v-a0c466dc><!--[--><!--]--><!----><span data-v-a0c466dc>Agent SDK</span><!--[--><!--[--><!--[--><!--[--><span class="agent-sdk-version" title="@cursor/july 0.1.111" data-v-97e8f22b>0.1.111</span><!--]--><!--]--><!--]--><!--]--></a></div></div><div class="content" data-v-b33015df><div class="content-body" data-v-b33015df><!--[--><!--]--><div class="VPNavBarSearch search" data-v-b33015df><!--[--><!----><div id="local-search"><button type="button" class="DocSearch DocSearch-Button" aria-label="Search"><span class="DocSearch-Button-Container"><span class="vp-icon DocSearch-Search-Icon"></span><span class="DocSearch-Button-Placeholder">Search</span></span><span class="DocSearch-Button-Keys"><kbd class="DocSearch-Button-Key"></kbd><kbd class="DocSearch-Button-Key">K</kbd></span></button></div><!--]--></div><nav aria-labelledby="main-nav-aria-label" class="VPNavBarMenu menu" data-v-b33015df data-v-8906c751><span id="main-nav-aria-label" class="visually-hidden" data-v-8906c751> Main Navigation </span><!--[--><!--[--><a class="VPLink link VPNavBarMenuLink" href="/docs/quickstart.html" tabindex="0" data-v-8906c751 data-v-e9bcc08e><!--[--><span data-v-e9bcc08e>Quickstart</span><!--]--></a><!--]--><!--[--><a class="VPLink link VPNavBarMenuLink" href="/docs/guides/mcp-oauth.html" tabindex="0" data-v-8906c751 data-v-e9bcc08e><!--[--><span data-v-e9bcc08e>Guides</span><!--]--></a><!--]--><!--[--><a class="VPLink link VPNavBarMenuLink" href="/docs/guides/github.html" tabindex="0" data-v-8906c751 data-v-e9bcc08e><!--[--><span data-v-e9bcc08e>Integrations</span><!--]--></a><!--]--><!--[--><a class="VPLink link VPNavBarMenuLink" href="/docs/templates/security-reviewer.html" tabindex="0" data-v-8906c751 data-v-e9bcc08e><!--[--><span data-v-e9bcc08e>Templates</span><!--]--></a><!--]--><!--[--><a class="VPLink link VPNavBarMenuLink" href="/docs/reference/project-layout.html" tabindex="0" data-v-8906c751 data-v-e9bcc08e><!--[--><span data-v-e9bcc08e>Reference</span><!--]--></a><!--]--><!--]--></nav><!----><div class="VPNavBarAppearance appearance" data-v-b33015df data-v-959ed110><button class="VPSwitch VPSwitchAppearance" type="button" role="switch" title aria-checked="false" data-v-959ed110 data-v-79ad6d7f data-v-0a895de9><span class="check" data-v-0a895de9><span class="icon" data-v-0a895de9><!--[--><span class="vpi-sun sun" data-v-79ad6d7f></span><span class="vpi-moon moon" data-v-79ad6d7f></span><!--]--></span></span></button></div><!----><div class="VPFlyout VPNavBarExtra extra" data-v-b33015df data-v-a713f501 data-v-1d7aa189><button type="button" class="button" aria-haspopup="true" aria-expanded="false" aria-label="extra navigation" data-v-1d7aa189><span class="vpi-more-horizontal icon" data-v-1d7aa189></span></button><div class="menu" data-v-1d7aa189><div class="VPMenu" data-v-1d7aa189 data-v-e34b7e14><!----><!--[--><!--[--><!----><div class="group" data-v-a713f501><div class="item appearance" data-v-a713f501><p class="label" data-v-a713f501>Appearance</p><div class="appearance-action" data-v-a713f501><button class="VPSwitch VPSwitchAppearance" type="button" role="switch" title aria-checked="false" data-v-a713f501 data-v-79ad6d7f data-v-0a895de9><span class="check" data-v-0a895de9><span class="icon" data-v-0a895de9><!--[--><span class="vpi-sun sun" data-v-79ad6d7f></span><span class="vpi-moon moon" data-v-79ad6d7f></span><!--]--></span></span></button></div></div></div><!----><!--]--><!--]--></div></div></div><!--[--><!--]--><button type="button" class="VPNavBarHamburger hamburger" aria-label="mobile navigation" aria-expanded="false" aria-controls="VPNavScreen" data-v-b33015df data-v-cb5d0edb><span class="container" data-v-cb5d0edb><span class="top" data-v-cb5d0edb></span><span class="middle" data-v-cb5d0edb></span><span class="bottom" data-v-cb5d0edb></span></span></button></div></div></div></div><div class="divider" data-v-b33015df><div class="divider-line" data-v-b33015df></div></div></div><!----></header><div class="VPLocalNav has-sidebar empty" data-v-76086e96 data-v-6f794310><div class="container" data-v-6f794310><button class="menu" aria-expanded="false" aria-controls="VPSidebarNav" data-v-6f794310><span class="vpi-align-left menu-icon" data-v-6f794310></span><span class="menu-text" data-v-6f794310>Menu</span></button><div class="VPLocalNavOutlineDropdown" style="--vp-vh:0px;" data-v-6f794310 data-v-bab84d9c><button data-v-bab84d9c>Return to top</button><!----></div></div></div><aside class="VPSidebar" data-v-76086e96 data-v-4daa293d><div class="curtain" data-v-4daa293d></div><nav class="nav" id="VPSidebarNav" aria-labelledby="sidebar-aria-label" tabindex="-1" data-v-4daa293d><span class="visually-hidden" id="sidebar-aria-label" data-v-4daa293d> Sidebar Navigation </span><!--[--><!--]--><!--[--><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Getting started</h2><!----></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Overview</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/quickstart.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Building your first agent</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0 collapsible" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Integrations</h2><div class="caret" role="button" aria-label="toggle section" tabindex="0" data-v-d79cbf7e><span class="vpi-chevron-right caret-icon" data-v-d79cbf7e></span></div></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/github.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>GitHub</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/slack.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Slack</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/webhooks.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Webhooks</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/convert-automation.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Cursor Automations</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/opentelemetry.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>OpenTelemetry</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/gitlab.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>GitLab</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/bitbucket.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Bitbucket</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0 collapsible" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Extensions</h2><div class="caret" role="button" aria-label="toggle section" tabindex="0" data-v-d79cbf7e><span class="vpi-chevron-right caret-icon" data-v-d79cbf7e></span></div></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/extensions.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Overview</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/cloud-agents.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Cloud Agents</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/grokbot-agents.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Grok Bot Agents</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/improve.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Self-improving Agents</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0 collapsible has-active" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Guides</h2><div class="caret" role="button" aria-label="toggle section" tabindex="0" data-v-d79cbf7e><span class="vpi-chevron-right caret-icon" data-v-d79cbf7e></span></div></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/mcp-oauth.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>MCP</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/agent-to-agent.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Agent-to-agent</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/deployment.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Deployment</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/evals.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Evals</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/hillclimbing.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Hillclimbing</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/building-with-agents.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Self Improving Agents</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/troubleshooting.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Troubleshooting</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0 collapsible" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Templates</h2><div class="caret" role="button" aria-label="toggle section" tabindex="0" data-v-d79cbf7e><span class="vpi-chevron-right caret-icon" data-v-d79cbf7e></span></div></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/templates/security-reviewer.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Security reviewer</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/templates/agentic-owners.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Agentic Approver</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/templates/pr-autofixer.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>PR autofixer</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/templates/triage.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Triage agent</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0 collapsible" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Reference</h2><div class="caret" role="button" aria-label="toggle section" tabindex="0" data-v-d79cbf7e><span class="vpi-chevron-right caret-icon" data-v-d79cbf7e></span></div></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/project-layout.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Project layout</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/agent-config.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Agent config</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/instructions.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Instructions</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/tools.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Tools</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/prompt.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>prompt</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/skills.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Skills</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/connections.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>MCP Connections</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/subagents.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Subagents</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/channels.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Channels</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/schedules.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Schedules & reminders</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/hooks.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Hooks</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/artifacts.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Artifacts</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/sessions.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Sessions & streaming</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/playground.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Playground</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/cli.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>CLI</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/http-api.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>HTTP API</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><!--]--><!--[--><!--]--></nav></aside><div class="VPContent has-sidebar" id="VPContent" data-v-76086e96 data-v-258cbba4><div class="VPDoc has-sidebar has-aside" data-v-258cbba4 data-v-afe580da><!--[--><!--]--><div class="container" data-v-afe580da><div class="aside" data-v-afe580da><div class="aside-curtain" data-v-afe580da></div><div class="aside-container" data-v-afe580da><div class="aside-content" data-v-afe580da><div class="VPDocAside" data-v-afe580da data-v-fb5275a4><!--[--><!--]--><!--[--><!--]--><nav aria-labelledby="doc-outline-aria-label" class="VPDocAsideOutline" data-v-fb5275a4 data-v-b3783f18><div class="content" data-v-b3783f18><div class="outline-marker" data-v-b3783f18></div><div aria-level="2" class="outline-title" id="doc-outline-aria-label" role="heading" data-v-b3783f18>On this page</div><ul class="VPDocOutlineItem root" data-v-b3783f18 data-v-72702c6e><!--[--><!--]--></ul></div></nav><!--[--><!--]--><div class="spacer" data-v-fb5275a4></div><!--[--><!--]--><!----><!--[--><!--]--><!--[--><!--]--></div></div></div></div><div class="content" data-v-afe580da><div class="content-container" data-v-afe580da><!--[--><!--]--><main class="main" data-v-afe580da><div style="position:relative;" class="vp-doc _docs_hillclimbing" data-v-afe580da><div><h1 id="hillclimbing" tabindex="-1">Hillclimbing <a class="header-anchor" href="#hillclimbing" aria-label="Permalink to "Hillclimbing""></a></h1><p>Make an agent better on fixed inputs: measure, change one lever, remeasure, and lock every kept win with an eval.</p><h2 id="what-is-hillclimbing" tabindex="-1">What is hillclimbing? <a class="header-anchor" href="#what-is-hillclimbing" aria-label="Permalink to "What is hillclimbing?""></a></h2><p>Hillclimbing is a measured improvement loop. You pin a few fixtures, name the one dominant problem in the run, change one lever, and check the same fixtures again. Keep only what helps. Every kept change lands an <a href="./evals.html">eval</a> so the win stays put.</p><p>You don't have to run the loop alone. The package ships a coding-agent skill that drives it with you.</p><div class="mermaid"></div><h2 id="how-do-i-hillclimb-an-agent-with-a-coding-agent" tabindex="-1">How do I hillclimb an agent with a coding agent? <a class="header-anchor" href="#how-do-i-hillclimb-an-agent-with-a-coding-agent" aria-label="Permalink to "How do I hillclimb an agent with a coding agent?""></a></h2><p>Have Cursor read <a href="./../skills/hillclimb/SKILL.html"><code>skills/hillclimb/SKILL.md</code></a>.</p><p>Tell it:</p><ol><li><strong>Which agent</strong> you're improving (path or slug)</li><li><strong>One to three fixtures</strong> you'll reuse every round: a PR URL, a saved webhook body, or a canonical chat prompt</li><li><strong>What "better" means</strong> this round: correct tool choice, fewer tools, lower latency, or output quality. Name the freeze line too: API shape, public output, and existing evals that must stay green</li></ol><p>The skill serves the agent, hits your fixtures, reads the session trajectory, proposes one change, remeasures, and checks with you before the next round.</p><p>Other skills cover the edges:</p><table tabindex="0"><thead><tr><th>When you need…</th><th>Skill</th></tr></thead><tbody><tr><td>The measured improvement loop</td><td><a href="./../skills/hillclimb/SKILL.html"><code>skills/hillclimb/SKILL.md</code></a></td></tr><tr><td>An eval that locks a kept win</td><td><a href="./../skills/evals/SKILL.html"><code>skills/evals/SKILL.md</code></a></td></tr><tr><td>Repeatable GitHub webhook inputs</td><td><a href="./../skills/github/SKILL.html"><code>skills/github/SKILL.md</code></a></td></tr><tr><td>A run that misbehaves</td><td><a href="./../skills/debug/SKILL.html"><code>skills/debug/SKILL.md</code></a></td></tr></tbody></table><p>See <a href="./building-with-agents.html">Building agents with agents</a> for every framework skill and a good first prompt.</p><h2 id="what-do-i-need-before-a-hillclimb-round" tabindex="-1">What do I need before a hillclimb round? <a class="header-anchor" href="#what-do-i-need-before-a-hillclimb-round" aria-label="Permalink to "What do I need before a hillclimb round?""></a></h2><p>Agree on four things before you edit:</p><ol><li><strong>The target agent</strong>: the project you're improving</li><li><strong>Fixtures</strong>: one to three fixed inputs you can compare across runs</li><li><strong>Success criteria</strong>: what better means this round</li><li><strong>The freeze line</strong>: what must not change</li></ol><p>Pin the input first. A moving fixture is noise. For GitHub agents, use <code>agent-sdk github replay</code> (see the <a href="./guides/github.html">GitHub guide</a>). For a single tool without a model turn, use <code>agent-sdk call</code>. For a chat turn, use <code>agent-sdk run --dir . --message "…"</code>.</p><h2 id="how-do-i-run-one-hillclimb-round" tabindex="-1">How do I run one hillclimb round? <a class="header-anchor" href="#how-do-i-run-one-hillclimb-round" aria-label="Permalink to "How do I run one hillclimb round?""></a></h2><p><strong>Measure.</strong> Hit the agent the way a user would: playground, channel HTTP, or Slack in <code>--dev</code>. Or ask the hillclimb skill to do it. <code>agent-sdk run</code> returns a JSON trajectory and writes a trace under the project state directory.</p><p><strong>Reflect.</strong> Score the trajectory, not impressions. Was the answer right? Did the model thrash (too many tools, fat evidence, grep loops)? Did it invent work the host should have prepared? Name the single dominant problem for this round in one sentence. Example: "Full-file dumps trigger grep loops."</p><p><strong>Change one lever.</strong> Prefer the smallest change that addresses that problem:</p><ol><li>Host prep: seed what the model needs so it doesn't hunt</li><li>Evidence shape: trim or reorder artifacts</li><li>Instructions and skills: tighten the procedure</li><li>Tool surface: remove or gate tools that invite wandering</li><li>Framework changes: only when the agent can't express the fix</li></ol><p><strong>Remeasure.</strong> Same fixtures. Diff tools, wall time, and quality side by side. Keep the change only if the target metric improves and the freeze line holds.</p><h2 id="how-do-i-lock-a-hillclimb-improvement-with-an-eval" tabindex="-1">How do I lock a hillclimb improvement with an eval? <a class="header-anchor" href="#how-do-i-lock-a-hillclimb-improvement-with-an-eval" aria-label="Permalink to "How do I lock a hillclimb improvement with an eval?""></a></h2><p>Every kept change needs an eval that would have failed before the change: a tool-choice gate, a <code>maxToolCalls</code> bound, or an output-shape check. Run <code>agent-sdk eval --dir . --json</code> between rounds. Never weaken an existing gate to pass the round.</p><p>Details live in <a href="./evals.html#keep-improvements-with-regression-evals">Keep improvements with regression evals</a>. The evals skill will author the case with you.</p><h2 id="what-habits-help-hillclimbing-stay-reliable" tabindex="-1">What habits help hillclimbing stay reliable? <a class="header-anchor" href="#what-habits-help-hillclimbing-stay-reliable" aria-label="Permalink to "What habits help hillclimbing stay reliable?""></a></h2><ul><li>One problem per round. Don't bundle "trim evidence and rewrite instructions" unless you chose that on purpose.</li><li>Keep fixtures fixed until you deliberately need a harder case.</li><li>Separate host work from model tools when you blame latency. Moving deterministic prep onto the host is often the biggest win. In one PR reviewer, host-prepared evidence cut turns from about 8 minutes to about 1 minute.</li><li>Spot-check quality on at least one fixture against a known-good answer. Efficiency-only climbs quietly drop findings.</li><li>Treat <code>turn.failed</code> with <code>"turn interrupted"</code> as expected when a follow-up or stop preempted the turn.</li><li>Don't deploy, post to real surfaces, or weaken evals as part of a climb.</li></ul><h2 id="related" tabindex="-1">Related <a class="header-anchor" href="#related" aria-label="Permalink to "Related""></a></h2><ul><li><a href="./evals.html">Evals</a></li><li><a href="./building-with-agents.html">Building agents with agents</a></li><li><a href="./guides/github.html">GitHub guide</a></li><li><a href="./troubleshooting.html">Fix common agent problems</a></li></ul></div></div></main><footer class="VPDocFooter" data-v-afe580da data-v-feab96e0><!--[--><!--]--><!----><nav class="prev-next" aria-labelledby="doc-footer-aria-label" data-v-feab96e0><span class="visually-hidden" id="doc-footer-aria-label" data-v-feab96e0>Pager</span><div class="pager" data-v-feab96e0><a class="VPLink link pager-link prev" href="/docs/evals.html" data-v-feab96e0><!--[--><span class="desc" data-v-feab96e0>Previous page</span><span class="title" data-v-feab96e0>Evals</span><!--]--></a></div><div class="pager" data-v-feab96e0><a class="VPLink link pager-link next" href="/docs/building-with-agents.html" data-v-feab96e0><!--[--><span class="desc" data-v-feab96e0>Next page</span><span class="title" data-v-feab96e0>Self Improving Agents</span><!--]--></a></div></nav></footer><!--[--><!--]--></div></div></div><!--[--><!--]--></div></div><!----><!--[--><!--]--></div></div>
|
|
57
|
-
|
|
56
|
+
<div id="app"><div class="Layout" data-v-76086e96><!--[--><!--]--><!--[--><span tabindex="-1" data-v-97fb5898></span><a href="#VPContent" class="VPSkipLink visually-hidden" data-v-97fb5898>Skip to content</a><!--]--><!----><header class="VPNav" data-v-76086e96 data-v-781b1279><div class="VPNavBar" data-v-781b1279 data-v-b33015df><div class="wrapper" data-v-b33015df><div class="container" data-v-b33015df><div class="title" data-v-b33015df><div class="VPNavBarTitle has-sidebar" data-v-b33015df data-v-a0c466dc><a class="title" href="/docs/" data-v-a0c466dc><!--[--><!--]--><!----><span data-v-a0c466dc>Agent SDK</span><!--[--><!--[--><!--[--><!--[--><span class="agent-sdk-version" title="@cursor/july 0.1.113" data-v-97e8f22b>0.1.113</span><!--]--><!--]--><!--]--><!--]--></a></div></div><div class="content" data-v-b33015df><div class="content-body" data-v-b33015df><!--[--><!--]--><div class="VPNavBarSearch search" data-v-b33015df><!--[--><!----><div id="local-search"><button type="button" class="DocSearch DocSearch-Button" aria-label="Search"><span class="DocSearch-Button-Container"><span class="vp-icon DocSearch-Search-Icon"></span><span class="DocSearch-Button-Placeholder">Search</span></span><span class="DocSearch-Button-Keys"><kbd class="DocSearch-Button-Key"></kbd><kbd class="DocSearch-Button-Key">K</kbd></span></button></div><!--]--></div><nav aria-labelledby="main-nav-aria-label" class="VPNavBarMenu menu" data-v-b33015df data-v-8906c751><span id="main-nav-aria-label" class="visually-hidden" data-v-8906c751> Main Navigation </span><!--[--><!--[--><a class="VPLink link VPNavBarMenuLink" href="/docs/quickstart.html" tabindex="0" data-v-8906c751 data-v-e9bcc08e><!--[--><span data-v-e9bcc08e>Quickstart</span><!--]--></a><!--]--><!--[--><a class="VPLink link VPNavBarMenuLink" href="/docs/guides/mcp-oauth.html" tabindex="0" data-v-8906c751 data-v-e9bcc08e><!--[--><span data-v-e9bcc08e>Guides</span><!--]--></a><!--]--><!--[--><a class="VPLink link VPNavBarMenuLink" href="/docs/guides/github.html" tabindex="0" data-v-8906c751 data-v-e9bcc08e><!--[--><span data-v-e9bcc08e>Integrations</span><!--]--></a><!--]--><!--[--><a class="VPLink link VPNavBarMenuLink" href="/docs/templates/security-reviewer.html" tabindex="0" data-v-8906c751 data-v-e9bcc08e><!--[--><span data-v-e9bcc08e>Templates</span><!--]--></a><!--]--><!--[--><a class="VPLink link VPNavBarMenuLink" href="/docs/reference/project-layout.html" tabindex="0" data-v-8906c751 data-v-e9bcc08e><!--[--><span data-v-e9bcc08e>Reference</span><!--]--></a><!--]--><!--]--></nav><!----><div class="VPNavBarAppearance appearance" data-v-b33015df data-v-959ed110><button class="VPSwitch VPSwitchAppearance" type="button" role="switch" title aria-checked="false" data-v-959ed110 data-v-79ad6d7f data-v-0a895de9><span class="check" data-v-0a895de9><span class="icon" data-v-0a895de9><!--[--><span class="vpi-sun sun" data-v-79ad6d7f></span><span class="vpi-moon moon" data-v-79ad6d7f></span><!--]--></span></span></button></div><!----><div class="VPFlyout VPNavBarExtra extra" data-v-b33015df data-v-a713f501 data-v-1d7aa189><button type="button" class="button" aria-haspopup="true" aria-expanded="false" aria-label="extra navigation" data-v-1d7aa189><span class="vpi-more-horizontal icon" data-v-1d7aa189></span></button><div class="menu" data-v-1d7aa189><div class="VPMenu" data-v-1d7aa189 data-v-e34b7e14><!----><!--[--><!--[--><!----><div class="group" data-v-a713f501><div class="item appearance" data-v-a713f501><p class="label" data-v-a713f501>Appearance</p><div class="appearance-action" data-v-a713f501><button class="VPSwitch VPSwitchAppearance" type="button" role="switch" title aria-checked="false" data-v-a713f501 data-v-79ad6d7f data-v-0a895de9><span class="check" data-v-0a895de9><span class="icon" data-v-0a895de9><!--[--><span class="vpi-sun sun" data-v-79ad6d7f></span><span class="vpi-moon moon" data-v-79ad6d7f></span><!--]--></span></span></button></div></div></div><!----><!--]--><!--]--></div></div></div><!--[--><!--]--><button type="button" class="VPNavBarHamburger hamburger" aria-label="mobile navigation" aria-expanded="false" aria-controls="VPNavScreen" data-v-b33015df data-v-cb5d0edb><span class="container" data-v-cb5d0edb><span class="top" data-v-cb5d0edb></span><span class="middle" data-v-cb5d0edb></span><span class="bottom" data-v-cb5d0edb></span></span></button></div></div></div></div><div class="divider" data-v-b33015df><div class="divider-line" data-v-b33015df></div></div></div><!----></header><div class="VPLocalNav has-sidebar empty" data-v-76086e96 data-v-6f794310><div class="container" data-v-6f794310><button class="menu" aria-expanded="false" aria-controls="VPSidebarNav" data-v-6f794310><span class="vpi-align-left menu-icon" data-v-6f794310></span><span class="menu-text" data-v-6f794310>Menu</span></button><div class="VPLocalNavOutlineDropdown" style="--vp-vh:0px;" data-v-6f794310 data-v-bab84d9c><button data-v-bab84d9c>Return to top</button><!----></div></div></div><aside class="VPSidebar" data-v-76086e96 data-v-4daa293d><div class="curtain" data-v-4daa293d></div><nav class="nav" id="VPSidebarNav" aria-labelledby="sidebar-aria-label" tabindex="-1" data-v-4daa293d><span class="visually-hidden" id="sidebar-aria-label" data-v-4daa293d> Sidebar Navigation </span><!--[--><!--]--><!--[--><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Getting started</h2><!----></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Overview</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/quickstart.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Building your first agent</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0 collapsible" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Integrations</h2><div class="caret" role="button" aria-label="toggle section" tabindex="0" data-v-d79cbf7e><span class="vpi-chevron-right caret-icon" data-v-d79cbf7e></span></div></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/github.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>GitHub</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/slack.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Slack</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/webhooks.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Webhooks</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/convert-automation.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Cursor Automations</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/opentelemetry.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>OpenTelemetry</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/gitlab.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>GitLab</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/bitbucket.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Bitbucket</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0 collapsible" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Extensions</h2><div class="caret" role="button" aria-label="toggle section" tabindex="0" data-v-d79cbf7e><span class="vpi-chevron-right caret-icon" data-v-d79cbf7e></span></div></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/extensions.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Overview</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/cloud-agents.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Cloud agents</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/grokbot-agents.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Grok Bot agents</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/improve.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Self-improvement</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/jev.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Jev</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0 collapsible has-active" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Guides</h2><div class="caret" role="button" aria-label="toggle section" tabindex="0" data-v-d79cbf7e><span class="vpi-chevron-right caret-icon" data-v-d79cbf7e></span></div></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/mcp-oauth.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>MCP</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/guides/agent-to-agent.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Peer agents</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/deployment.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Deployment</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/evals.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Evals</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/hillclimbing.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Hillclimbing</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/building-with-agents.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Build with a coding agent</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/troubleshooting.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Troubleshooting</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0 collapsible" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Templates</h2><div class="caret" role="button" aria-label="toggle section" tabindex="0" data-v-d79cbf7e><span class="vpi-chevron-right caret-icon" data-v-d79cbf7e></span></div></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/templates/security-reviewer.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Security reviewer</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/templates/thermo-review.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Thermo review</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/templates/thermo-quality-review.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Thermo quality review</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/templates/agentic-owners.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Agentic Approver</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/templates/pr-autofixer.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>PR autofixer</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/templates/triage.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Triage agent</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><div class="no-transition group" data-v-095feacd><section class="VPSidebarItem level-0 collapsible" data-v-095feacd data-v-d79cbf7e><div class="item" role="button" tabindex="0" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><h2 class="text" data-v-d79cbf7e>Reference</h2><div class="caret" role="button" aria-label="toggle section" tabindex="0" data-v-d79cbf7e><span class="vpi-chevron-right caret-icon" data-v-d79cbf7e></span></div></div><div class="items" data-v-d79cbf7e><!--[--><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/project-layout.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Project layout</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/agent-config.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Agent config</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/instructions.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Instructions</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/tools.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Tools</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/prompt.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>prompt</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/skills.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Skills</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/connections.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>MCP Connections</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/subagents.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Subagents</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/channels.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Channels</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/schedules.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Schedules & reminders</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/hooks.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Hooks</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/evals.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Evals</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/artifacts.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Artifacts</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/sessions.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Sessions & streaming</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/playground.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>Playground</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/cli.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>CLI</p><!--]--></a><!----></div><!----></div><div class="VPSidebarItem level-1 is-link" data-v-d79cbf7e><div class="item" data-v-d79cbf7e><div class="indicator" data-v-d79cbf7e></div><a class="VPLink link link" href="/docs/reference/http-api.html" data-v-d79cbf7e><!--[--><p class="text" data-v-d79cbf7e>HTTP API</p><!--]--></a><!----></div><!----></div><!--]--></div></section></div><!--]--><!--[--><!--]--></nav></aside><div class="VPContent has-sidebar" id="VPContent" data-v-76086e96 data-v-258cbba4><div class="VPDoc has-sidebar has-aside" data-v-258cbba4 data-v-afe580da><!--[--><!--]--><div class="container" data-v-afe580da><div class="aside" data-v-afe580da><div class="aside-curtain" data-v-afe580da></div><div class="aside-container" data-v-afe580da><div class="aside-content" data-v-afe580da><div class="VPDocAside" data-v-afe580da data-v-fb5275a4><!--[--><!--]--><!--[--><!--]--><nav aria-labelledby="doc-outline-aria-label" class="VPDocAsideOutline" data-v-fb5275a4 data-v-b3783f18><div class="content" data-v-b3783f18><div class="outline-marker" data-v-b3783f18></div><div aria-level="2" class="outline-title" id="doc-outline-aria-label" role="heading" data-v-b3783f18>On this page</div><ul class="VPDocOutlineItem root" data-v-b3783f18 data-v-72702c6e><!--[--><!--]--></ul></div></nav><!--[--><!--]--><div class="spacer" data-v-fb5275a4></div><!--[--><!--]--><!----><!--[--><!--]--><!--[--><!--]--></div></div></div></div><div class="content" data-v-afe580da><div class="content-container" data-v-afe580da><!--[--><!--]--><main class="main" data-v-afe580da><div style="position:relative;" class="vp-doc _docs_hillclimbing" data-v-afe580da><div><h1 id="hillclimbing" tabindex="-1">Hillclimbing <a class="header-anchor" href="#hillclimbing" aria-label="Permalink to "Hillclimbing""></a></h1><p>Hillclimbing improves an agent through controlled before-and-after comparisons. Keep the inputs fixed, name one failure, change one lever, and retain the change only when the same run gets better without crossing the freeze line.</p><p>The companion <a href="./../skills/hillclimb/SKILL.html"><code>hillclimb</code> skill</a> can run this loop with you.</p><h2 id="correct-a-wrong-tool-choice" tabindex="-1">Correct a wrong tool choice <a class="header-anchor" href="#correct-a-wrong-tool-choice" aria-label="Permalink to "Correct a wrong tool choice""></a></h2><p>Suppose a PR reviewer calls <code>approve_pr</code> before <code>inspect_pr</code>. Pin one PR fixture, record the original tool order, then tighten the instruction that owns the decision. The improved run should inspect first and leave approval untouched.</p><div class="language-bash vp-adaptive-theme"><button title="Copy Code" class="copy"></button><span class="lang">bash</span><pre class="shiki shiki-themes github-light github-dark vp-code" tabindex="0"><code><span class="line"><span style="--shiki-light:#6F42C1;--shiki-dark:#B392F0;">agent-sdk</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> run</span><span style="--shiki-light:#005CC5;--shiki-dark:#79B8FF;"> --dir</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> .</span><span style="--shiki-light:#005CC5;--shiki-dark:#79B8FF;"> \</span></span>
|
|
57
|
+
<span class="line"><span style="--shiki-light:#005CC5;--shiki-dark:#79B8FF;"> --message</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> "Is https://github.com/acme/checkout/pull/42 ready?"</span></span></code></pre></div><p>Lock the result with the intended and tempting-wrong paths:</p><div class="language-ts vp-adaptive-theme"><button title="Copy Code" class="copy"></button><span class="lang">ts</span><pre class="shiki shiki-themes github-light github-dark vp-code" tabindex="0"><code><span class="line"><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">t.</span><span style="--shiki-light:#6F42C1;--shiki-dark:#B392F0;">succeeded</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">();</span></span>
|
|
58
|
+
<span class="line"><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">t.</span><span style="--shiki-light:#6F42C1;--shiki-dark:#B392F0;">calledTool</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">(</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;">"inspect_pr"</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">);</span></span>
|
|
59
|
+
<span class="line"><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">t.</span><span style="--shiki-light:#6F42C1;--shiki-dark:#B392F0;">notCalledTool</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">(</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;">"approve_pr"</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">);</span></span></code></pre></div><p>Do not change the tool implementation, prompt, and evidence together. If tool choice improves, you should know which edit caused it.</p><h2 id="reduce-wandering-or-latency" tabindex="-1">Reduce wandering or latency <a class="header-anchor" href="#reduce-wandering-or-latency" aria-label="Permalink to "Reduce wandering or latency""></a></h2><p>A slow run often repeats reads, searches broad directories, or asks the model to discover evidence the host already has. Compare the trajectory by tool name and wall time, then move deterministic preparation ahead of the turn or trim the evidence shape.</p><div class="language-text vp-adaptive-theme"><button title="Copy Code" class="copy"></button><span class="lang">text</span><pre class="shiki shiki-themes github-light github-dark vp-code" tabindex="0"><code><span class="line"><span>before: 14 tool calls, 92 seconds, 6 repeated reads</span></span>
|
|
60
|
+
<span class="line"><span>after: 5 tool calls, 31 seconds, 0 repeated reads</span></span></code></pre></div><p>Keep the change only if the answer still finds the same known issue. Efficiency without a quality check can make an agent fast by making it incomplete.</p><h2 id="improve-quality-without-changing-the-contract" tabindex="-1">Improve quality without changing the contract <a class="header-anchor" href="#improve-quality-without-changing-the-contract" aria-label="Permalink to "Improve quality without changing the contract""></a></h2><p>For answer quality, freeze the public shape before editing: required sections, API fields, side effects, and existing evals. Compare both runs against a known-good answer or a small set of gold findings, then change the instructions or skill that owns the missing judgment.</p><div class="language-text vp-adaptive-theme"><button title="Copy Code" class="copy"></button><span class="lang">text</span><pre class="shiki shiki-themes github-light github-dark vp-code" tabindex="0"><code><span class="line"><span>fixture: saved checkout PR diff</span></span>
|
|
61
|
+
<span class="line"><span>success: report both gold findings with file paths</span></span>
|
|
62
|
+
<span class="line"><span>freeze: findings JSON shape; no review post; existing smoke evals</span></span></code></pre></div><p>A wording change is not a win by itself. Keep it when correctness rises and every frozen behavior remains intact.</p><h2 id="run-one-measured-round" tabindex="-1">Run one measured round <a class="header-anchor" href="#run-one-measured-round" aria-label="Permalink to "Run one measured round""></a></h2><p>Agree on four inputs before editing:</p><ol><li><strong>Target:</strong> the agent path or slug.</li><li><strong>Fixtures:</strong> one to three prompts, saved payloads, or pinned PRs.</li><li><strong>Success:</strong> the metric or decision that should improve.</li><li><strong>Freeze line:</strong> behavior that must not move.</li></ol><p>Then run one loop:</p><ol><li>Exercise the agent through the same surface a user would.</li><li>Read the trajectory and name the dominant failure in one sentence.</li><li>Change one lever.</li><li>Rerun the same fixtures.</li><li>Compare correctness, efficiency, and contract fit.</li><li>Keep, revert, or narrow the change.</li></ol><p>Do not replace a fixture mid-round. A moving input makes the comparison meaningless.</p><h2 id="choose-the-smallest-lever" tabindex="-1">Choose the smallest lever <a class="header-anchor" href="#choose-the-smallest-lever" aria-label="Permalink to "Choose the smallest lever""></a></h2><p>Start with the earliest place that can remove the failure:</p><ol><li><strong>Host preparation:</strong> fetch deterministic evidence before the turn.</li><li><strong>Evidence shape:</strong> trim, order, or label what the model receives.</li><li><strong>Instructions or skills:</strong> clarify the decision procedure.</li><li><strong>Tool surface:</strong> remove or gate tools that invite wandering.</li><li><strong>Framework:</strong> change it only when the agent cannot express the fix.</li></ol><p>Changing the host or evidence often improves both quality and latency because the model spends less time searching for ground truth.</p><h2 id="lock-every-kept-win" tabindex="-1">Lock every kept win <a class="header-anchor" href="#lock-every-kept-win" aria-label="Permalink to "Lock every kept win""></a></h2><p>Each retained change needs an eval that would have failed before it. Use a tool-choice assertion, <code>maxToolCalls</code> bound, output-shape check, or quality score that captures the measured improvement.</p><div class="language-bash vp-adaptive-theme"><button title="Copy Code" class="copy"></button><span class="lang">bash</span><pre class="shiki shiki-themes github-light github-dark vp-code" tabindex="0"><code><span class="line"><span style="--shiki-light:#6F42C1;--shiki-dark:#B392F0;">agent-sdk</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> eval</span><span style="--shiki-light:#005CC5;--shiki-dark:#79B8FF;"> --dir</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> .</span><span style="--shiki-light:#005CC5;--shiki-dark:#79B8FF;"> --tag</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> smoke</span></span></code></pre></div><p>Never weaken an existing gate to make the round pass. The eval is the record that the improvement survived after the fixture and source change leave your working memory.</p><h2 id="run-the-loop-with-a-coding-agent" tabindex="-1">Run the loop with a coding agent <a class="header-anchor" href="#run-the-loop-with-a-coding-agent" aria-label="Permalink to "Run the loop with a coding agent""></a></h2><p>Ask the coding agent to read <a href="./../skills/hillclimb/SKILL.html"><code>skills/hillclimb/SKILL.md</code></a>, then give it the target, fixtures, success criterion, and freeze line. It will measure the current run, propose one change, remeasure, and ask before starting another round.</p><p>Use saved or local surfaces during a climb. Do not deploy or post to a real external channel unless that side effect is part of an explicitly approved fixture.</p><h2 id="related" tabindex="-1">Related <a class="header-anchor" href="#related" aria-label="Permalink to "Related""></a></h2><ul><li><a href="./evals.html">Evals</a>: author the regression check for a kept win</li><li><a href="./reference/evals.html">Evals reference</a>: trajectory assertions and scores</li><li><a href="./building-with-agents.html">Building agents with agents</a>: delegate creation and verification to a coding agent</li><li><a href="./troubleshooting.html">Fix common agent problems</a>: diagnose a run that is broken instead of underperforming</li></ul></div></div></main><footer class="VPDocFooter" data-v-afe580da data-v-feab96e0><!--[--><!--]--><!----><nav class="prev-next" aria-labelledby="doc-footer-aria-label" data-v-feab96e0><span class="visually-hidden" id="doc-footer-aria-label" data-v-feab96e0>Pager</span><div class="pager" data-v-feab96e0><a class="VPLink link pager-link prev" href="/docs/evals.html" data-v-feab96e0><!--[--><span class="desc" data-v-feab96e0>Previous page</span><span class="title" data-v-feab96e0>Evals</span><!--]--></a></div><div class="pager" data-v-feab96e0><a class="VPLink link pager-link next" href="/docs/building-with-agents.html" data-v-feab96e0><!--[--><span class="desc" data-v-feab96e0>Next page</span><span class="title" data-v-feab96e0>Build with a coding agent</span><!--]--></a></div></nav></footer><!--[--><!--]--></div></div></div><!--[--><!--]--></div></div><!----><!--[--><!--]--></div></div>
|
|
63
|
+
<script>window.__VP_HASH_MAP__=JSON.parse("{\"building-with-agents.md\":\"BBCx0AUo\",\"deployment.md\":\"D2jQZuFx\",\"evals.md\":\"D3Y3Aixt\",\"guides_agent-to-agent.md\":\"CD4T5FIl\",\"guides_bitbucket.md\":\"mpevW-VP\",\"guides_cloud-agents.md\":\"Cp1O3u-X\",\"guides_convert-automation.md\":\"CqEyfP6Y\",\"guides_github.md\":\"BwpBp3ed\",\"guides_gitlab.md\":\"DaEC3nMk\",\"guides_grokbot-agents.md\":\"CMhZNdEU\",\"guides_improve.md\":\"Bnp4F99w\",\"guides_jev.md\":\"F5fAkkfN\",\"guides_mcp-oauth.md\":\"bSFakfCY\",\"guides_opentelemetry.md\":\"BKDxQmmd\",\"guides_slack.md\":\"Bo96y42E\",\"guides_webhooks.md\":\"1A72_VEE\",\"hillclimbing.md\":\"D4E1o5Sa\",\"index.md\":\"C-t81M5J\",\"quickstart.md\":\"DAvVhuuU\",\"reference_agent-config.md\":\"DGPyw7ms\",\"reference_artifacts.md\":\"Bu_4HmsD\",\"reference_channels.md\":\"nFWbzAic\",\"reference_cli.md\":\"DLWDz9ij\",\"reference_connections.md\":\"Je9dMsdd\",\"reference_evals.md\":\"DNJzM_yf\",\"reference_extensions.md\":\"Cv5aLCz_\",\"reference_hooks.md\":\"B7uzNENk\",\"reference_http-api.md\":\"CduHavZ2\",\"reference_instructions.md\":\"CU1My5My\",\"reference_playground.md\":\"Ch2d0Iqi\",\"reference_project-layout.md\":\"BGhgpy9V\",\"reference_prompt.md\":\"Ccp0R53H\",\"reference_schedules.md\":\"B2Nm6FaD\",\"reference_sessions.md\":\"1_6Vyv7x\",\"reference_skills.md\":\"DjQkRefx\",\"reference_subagents.md\":\"Dl16gcBj\",\"reference_tools.md\":\"B1dH1lpa\",\"templates_agentic-owners.md\":\"9M575F5C\",\"templates_pr-autofixer.md\":\"ws0DDXDy\",\"templates_security-reviewer.md\":\"KEFYXzfK\",\"templates_thermo-quality-review.md\":\"VNJ_mohX\",\"templates_thermo-review.md\":\"Hi3zWOkP\",\"templates_triage.md\":\"CVGe_FG6\",\"troubleshooting.md\":\"mnfFG2Em\"}");window.__VP_SITE_DATA__=JSON.parse("{\"lang\":\"en-US\",\"dir\":\"ltr\",\"title\":\"Agent SDK\",\"description\":\"Filesystem-first framework for building and serving Cursor agents.\",\"base\":\"/docs/\",\"head\":[],\"router\":{\"prefetchLinks\":true},\"appearance\":true,\"themeConfig\":{\"sdkVersion\":\"0.1.113\",\"nav\":[{\"text\":\"Quickstart\",\"link\":\"/quickstart\"},{\"text\":\"Guides\",\"link\":\"/guides/mcp-oauth\"},{\"text\":\"Integrations\",\"link\":\"/guides/github\"},{\"text\":\"Templates\",\"link\":\"/templates/security-reviewer\"},{\"text\":\"Reference\",\"link\":\"/reference/project-layout\"}],\"search\":{\"provider\":\"local\"},\"outline\":{\"level\":[2,3]},\"sidebar\":[{\"text\":\"Getting started\",\"items\":[{\"text\":\"Overview\",\"link\":\"/\"},{\"text\":\"Building your first agent\",\"link\":\"/quickstart\"}]},{\"text\":\"Integrations\",\"collapsed\":false,\"items\":[{\"text\":\"GitHub\",\"link\":\"/guides/github\"},{\"text\":\"Slack\",\"link\":\"/guides/slack\"},{\"text\":\"Webhooks\",\"link\":\"/guides/webhooks\"},{\"text\":\"Cursor Automations\",\"link\":\"/guides/convert-automation\"},{\"text\":\"OpenTelemetry\",\"link\":\"/guides/opentelemetry\"},{\"text\":\"GitLab\",\"link\":\"/guides/gitlab\"},{\"text\":\"Bitbucket\",\"link\":\"/guides/bitbucket\"}]},{\"text\":\"Extensions\",\"collapsed\":false,\"items\":[{\"text\":\"Overview\",\"link\":\"/reference/extensions\"},{\"text\":\"Cloud agents\",\"link\":\"/guides/cloud-agents\"},{\"text\":\"Grok Bot agents\",\"link\":\"/guides/grokbot-agents\"},{\"text\":\"Self-improvement\",\"link\":\"/guides/improve\"},{\"text\":\"Jev\",\"link\":\"/guides/jev\"}]},{\"text\":\"Guides\",\"collapsed\":false,\"items\":[{\"text\":\"MCP\",\"link\":\"/guides/mcp-oauth\"},{\"text\":\"Peer agents\",\"link\":\"/guides/agent-to-agent\"},{\"text\":\"Deployment\",\"link\":\"/deployment\"},{\"text\":\"Evals\",\"link\":\"/evals\"},{\"text\":\"Hillclimbing\",\"link\":\"/hillclimbing\"},{\"text\":\"Build with a coding agent\",\"link\":\"/building-with-agents\"},{\"text\":\"Troubleshooting\",\"link\":\"/troubleshooting\"}]},{\"text\":\"Templates\",\"collapsed\":false,\"items\":[{\"text\":\"Security reviewer\",\"link\":\"/templates/security-reviewer\"},{\"text\":\"Thermo review\",\"link\":\"/templates/thermo-review\"},{\"text\":\"Thermo quality review\",\"link\":\"/templates/thermo-quality-review\"},{\"text\":\"Agentic Approver\",\"link\":\"/templates/agentic-owners\"},{\"text\":\"PR autofixer\",\"link\":\"/templates/pr-autofixer\"},{\"text\":\"Triage agent\",\"link\":\"/templates/triage\"}]},{\"text\":\"Reference\",\"collapsed\":false,\"items\":[{\"text\":\"Project layout\",\"link\":\"/reference/project-layout\"},{\"text\":\"Agent config\",\"link\":\"/reference/agent-config\"},{\"text\":\"Instructions\",\"link\":\"/reference/instructions\"},{\"text\":\"Tools\",\"link\":\"/reference/tools\"},{\"text\":\"prompt\",\"link\":\"/reference/prompt\"},{\"text\":\"Skills\",\"link\":\"/reference/skills\"},{\"text\":\"MCP Connections\",\"link\":\"/reference/connections\"},{\"text\":\"Subagents\",\"link\":\"/reference/subagents\"},{\"text\":\"Channels\",\"link\":\"/reference/channels\"},{\"text\":\"Schedules & reminders\",\"link\":\"/reference/schedules\"},{\"text\":\"Hooks\",\"link\":\"/reference/hooks\"},{\"text\":\"Evals\",\"link\":\"/reference/evals\"},{\"text\":\"Artifacts\",\"link\":\"/reference/artifacts\"},{\"text\":\"Sessions & streaming\",\"link\":\"/reference/sessions\"},{\"text\":\"Playground\",\"link\":\"/reference/playground\"},{\"text\":\"CLI\",\"link\":\"/reference/cli\"},{\"text\":\"HTTP API\",\"link\":\"/reference/http-api\"}]}]},\"locales\":{},\"scrollOffset\":134,\"cleanUrls\":false}");</script>
|
|
58
64
|
|
|
59
65
|
</body>
|
|
60
66
|
</html>
|
|
@@ -1,88 +1,134 @@
|
|
|
1
1
|
# Hillclimbing
|
|
2
2
|
|
|
3
|
-
|
|
3
|
+
Hillclimbing improves an agent through controlled before-and-after
|
|
4
|
+
comparisons. Keep the inputs fixed, name one failure, change one lever,
|
|
5
|
+
and retain the change only when the same run gets better without
|
|
6
|
+
crossing the freeze line.
|
|
4
7
|
|
|
5
|
-
|
|
8
|
+
The companion [`hillclimb` skill](https://github.com/cursor/cursor/blob/main/packages/agent-serve/skills/hillclimb/SKILL.md) can run
|
|
9
|
+
this loop with you.
|
|
6
10
|
|
|
7
|
-
|
|
11
|
+
## Correct a wrong tool choice
|
|
8
12
|
|
|
9
|
-
|
|
13
|
+
Suppose a PR reviewer calls `approve_pr` before `inspect_pr`. Pin one PR
|
|
14
|
+
fixture, record the original tool order, then tighten the instruction
|
|
15
|
+
that owns the decision. The improved run should inspect first and leave
|
|
16
|
+
approval untouched.
|
|
10
17
|
|
|
11
|
-
```
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
change --> remeasure[Remeasure]
|
|
15
|
-
remeasure --> measure
|
|
18
|
+
```bash
|
|
19
|
+
agent-sdk run --dir . \
|
|
20
|
+
--message "Is https://github.com/acme/checkout/pull/42 ready?"
|
|
16
21
|
```
|
|
17
22
|
|
|
18
|
-
|
|
23
|
+
Lock the result with the intended and tempting-wrong paths:
|
|
19
24
|
|
|
20
|
-
|
|
25
|
+
```ts
|
|
26
|
+
t.succeeded();
|
|
27
|
+
t.calledTool("inspect_pr");
|
|
28
|
+
t.notCalledTool("approve_pr");
|
|
29
|
+
```
|
|
30
|
+
|
|
31
|
+
Do not change the tool implementation, prompt, and evidence together.
|
|
32
|
+
If tool choice improves, you should know which edit caused it.
|
|
21
33
|
|
|
22
|
-
|
|
34
|
+
## Reduce wandering or latency
|
|
23
35
|
|
|
24
|
-
|
|
25
|
-
|
|
26
|
-
|
|
36
|
+
A slow run often repeats reads, searches broad directories, or asks the
|
|
37
|
+
model to discover evidence the host already has. Compare the trajectory
|
|
38
|
+
by tool name and wall time, then move deterministic preparation ahead
|
|
39
|
+
of the turn or trim the evidence shape.
|
|
27
40
|
|
|
28
|
-
|
|
41
|
+
```text
|
|
42
|
+
before: 14 tool calls, 92 seconds, 6 repeated reads
|
|
43
|
+
after: 5 tool calls, 31 seconds, 0 repeated reads
|
|
44
|
+
```
|
|
29
45
|
|
|
30
|
-
|
|
46
|
+
Keep the change only if the answer still finds the same known issue.
|
|
47
|
+
Efficiency without a quality check can make an agent fast by making it
|
|
48
|
+
incomplete.
|
|
31
49
|
|
|
32
|
-
|
|
33
|
-
| --- | --- |
|
|
34
|
-
| The measured improvement loop | [`skills/hillclimb/SKILL.md`](https://github.com/cursor/cursor/blob/main/packages/agent-serve/skills/hillclimb/SKILL.md) |
|
|
35
|
-
| An eval that locks a kept win | [`skills/evals/SKILL.md`](https://github.com/cursor/cursor/blob/main/packages/agent-serve/skills/evals/SKILL.md) |
|
|
36
|
-
| Repeatable GitHub webhook inputs | [`skills/github/SKILL.md`](https://github.com/cursor/cursor/blob/main/packages/agent-serve/skills/github/SKILL.md) |
|
|
37
|
-
| A run that misbehaves | [`skills/debug/SKILL.md`](https://github.com/cursor/cursor/blob/main/packages/agent-serve/skills/debug/SKILL.md) |
|
|
50
|
+
## Improve quality without changing the contract
|
|
38
51
|
|
|
39
|
-
|
|
52
|
+
For answer quality, freeze the public shape before editing: required
|
|
53
|
+
sections, API fields, side effects, and existing evals. Compare both
|
|
54
|
+
runs against a known-good answer or a small set of gold findings, then
|
|
55
|
+
change the instructions or skill that owns the missing judgment.
|
|
56
|
+
|
|
57
|
+
```text
|
|
58
|
+
fixture: saved checkout PR diff
|
|
59
|
+
success: report both gold findings with file paths
|
|
60
|
+
freeze: findings JSON shape; no review post; existing smoke evals
|
|
61
|
+
```
|
|
40
62
|
|
|
41
|
-
|
|
63
|
+
A wording change is not a win by itself. Keep it when correctness rises
|
|
64
|
+
and every frozen behavior remains intact.
|
|
42
65
|
|
|
43
|
-
|
|
66
|
+
## Run one measured round
|
|
44
67
|
|
|
45
|
-
|
|
46
|
-
2. **Fixtures**: one to three fixed inputs you can compare across runs
|
|
47
|
-
3. **Success criteria**: what better means this round
|
|
48
|
-
4. **The freeze line**: what must not change
|
|
68
|
+
Agree on four inputs before editing:
|
|
49
69
|
|
|
50
|
-
|
|
70
|
+
1. **Target:** the agent path or slug.
|
|
71
|
+
2. **Fixtures:** one to three prompts, saved payloads, or pinned PRs.
|
|
72
|
+
3. **Success:** the metric or decision that should improve.
|
|
73
|
+
4. **Freeze line:** behavior that must not move.
|
|
51
74
|
|
|
52
|
-
|
|
75
|
+
Then run one loop:
|
|
53
76
|
|
|
54
|
-
|
|
77
|
+
1. Exercise the agent through the same surface a user would.
|
|
78
|
+
2. Read the trajectory and name the dominant failure in one sentence.
|
|
79
|
+
3. Change one lever.
|
|
80
|
+
4. Rerun the same fixtures.
|
|
81
|
+
5. Compare correctness, efficiency, and contract fit.
|
|
82
|
+
6. Keep, revert, or narrow the change.
|
|
55
83
|
|
|
56
|
-
|
|
84
|
+
Do not replace a fixture mid-round. A moving input makes the comparison
|
|
85
|
+
meaningless.
|
|
57
86
|
|
|
58
|
-
|
|
87
|
+
## Choose the smallest lever
|
|
59
88
|
|
|
60
|
-
|
|
61
|
-
2. Evidence shape: trim or reorder artifacts
|
|
62
|
-
3. Instructions and skills: tighten the procedure
|
|
63
|
-
4. Tool surface: remove or gate tools that invite wandering
|
|
64
|
-
5. Framework changes: only when the agent can't express the fix
|
|
89
|
+
Start with the earliest place that can remove the failure:
|
|
65
90
|
|
|
66
|
-
|
|
91
|
+
1. **Host preparation:** fetch deterministic evidence before the turn.
|
|
92
|
+
2. **Evidence shape:** trim, order, or label what the model receives.
|
|
93
|
+
3. **Instructions or skills:** clarify the decision procedure.
|
|
94
|
+
4. **Tool surface:** remove or gate tools that invite wandering.
|
|
95
|
+
5. **Framework:** change it only when the agent cannot express the fix.
|
|
67
96
|
|
|
68
|
-
|
|
97
|
+
Changing the host or evidence often improves both quality and latency
|
|
98
|
+
because the model spends less time searching for ground truth.
|
|
99
|
+
|
|
100
|
+
## Lock every kept win
|
|
101
|
+
|
|
102
|
+
Each retained change needs an eval that would have failed before it. Use
|
|
103
|
+
a tool-choice assertion, `maxToolCalls` bound, output-shape check, or
|
|
104
|
+
quality score that captures the measured improvement.
|
|
105
|
+
|
|
106
|
+
```bash
|
|
107
|
+
agent-sdk eval --dir . --tag smoke
|
|
108
|
+
```
|
|
69
109
|
|
|
70
|
-
|
|
110
|
+
Never weaken an existing gate to make the round pass. The eval is the
|
|
111
|
+
record that the improvement survived after the fixture and source
|
|
112
|
+
change leave your working memory.
|
|
71
113
|
|
|
72
|
-
|
|
114
|
+
## Run the loop with a coding agent
|
|
73
115
|
|
|
74
|
-
|
|
116
|
+
Ask the coding agent to read
|
|
117
|
+
[`skills/hillclimb/SKILL.md`](https://github.com/cursor/cursor/blob/main/packages/agent-serve/skills/hillclimb/SKILL.md), then give
|
|
118
|
+
it the target, fixtures, success criterion, and freeze line. It will
|
|
119
|
+
measure the current run, propose one change, remeasure, and ask before
|
|
120
|
+
starting another round.
|
|
75
121
|
|
|
76
|
-
|
|
77
|
-
|
|
78
|
-
|
|
79
|
-
- Spot-check quality on at least one fixture against a known-good answer. Efficiency-only climbs quietly drop findings.
|
|
80
|
-
- Treat `turn.failed` with `"turn interrupted"` as expected when a follow-up or stop preempted the turn.
|
|
81
|
-
- Don't deploy, post to real surfaces, or weaken evals as part of a climb.
|
|
122
|
+
Use saved or local surfaces during a climb. Do not deploy or post to a
|
|
123
|
+
real external channel unless that side effect is part of an explicitly
|
|
124
|
+
approved fixture.
|
|
82
125
|
|
|
83
126
|
## Related
|
|
84
127
|
|
|
85
|
-
- [Evals](/docs/evals.md)
|
|
86
|
-
- [
|
|
87
|
-
|
|
88
|
-
- [
|
|
128
|
+
- [Evals](/docs/evals.md): author the regression check for a kept win
|
|
129
|
+
- [Evals reference](/docs/reference/evals.md): trajectory assertions and
|
|
130
|
+
scores
|
|
131
|
+
- [Building agents with agents](/docs/building-with-agents.md): delegate
|
|
132
|
+
creation and verification to a coding agent
|
|
133
|
+
- [Fix common agent problems](/docs/troubleshooting.md): diagnose a run
|
|
134
|
+
that is broken instead of underperforming
|