open-research-protocol 0.4.7 → 0.4.9

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.
Files changed (50) hide show
  1. package/README.md +15 -0
  2. package/cli/orp.py +1158 -43
  3. package/docs/AGENT_LOOP.md +3 -0
  4. package/docs/ORP_REASONING_KERNEL_AGENT_PILOT.md +125 -0
  5. package/docs/ORP_REASONING_KERNEL_AGENT_REPLICATION.md +97 -0
  6. package/docs/ORP_REASONING_KERNEL_CANONICAL_CONTINUATION_PILOT.md +100 -0
  7. package/docs/ORP_REASONING_KERNEL_COMPARISON_PILOT.md +116 -0
  8. package/docs/ORP_REASONING_KERNEL_CONTINUATION_PILOT.md +86 -0
  9. package/docs/ORP_REASONING_KERNEL_EVALUATION_PLAN.md +261 -0
  10. package/docs/ORP_REASONING_KERNEL_EVIDENCE_MATRIX.md +131 -0
  11. package/docs/ORP_REASONING_KERNEL_EVOLUTION.md +123 -0
  12. package/docs/ORP_REASONING_KERNEL_PICKUP_PILOT.md +107 -0
  13. package/docs/ORP_REASONING_KERNEL_TECHNICAL_VALIDATION.md +140 -22
  14. package/docs/ORP_REASONING_KERNEL_V0_1.md +11 -0
  15. package/docs/ORP_YOUTUBE_INSPECT.md +97 -0
  16. package/docs/benchmarks/orp_reasoning_kernel_agent_pilot_v0_1.json +796 -0
  17. package/docs/benchmarks/orp_reasoning_kernel_agent_replication_task_smoke.json +487 -0
  18. package/docs/benchmarks/orp_reasoning_kernel_agent_replication_v0_1.json +1927 -0
  19. package/docs/benchmarks/orp_reasoning_kernel_agent_replication_v0_2.json +10217 -0
  20. package/docs/benchmarks/orp_reasoning_kernel_canonical_continuation_task_smoke.json +174 -0
  21. package/docs/benchmarks/orp_reasoning_kernel_canonical_continuation_v0_1.json +598 -0
  22. package/docs/benchmarks/orp_reasoning_kernel_comparison_v0_1.json +688 -0
  23. package/docs/benchmarks/orp_reasoning_kernel_continuation_task_smoke.json +150 -0
  24. package/docs/benchmarks/orp_reasoning_kernel_continuation_v0_1.json +448 -0
  25. package/docs/benchmarks/orp_reasoning_kernel_pickup_v0_1.json +594 -0
  26. package/docs/benchmarks/orp_reasoning_kernel_v0_1_validation.json +769 -41
  27. package/examples/README.md +2 -0
  28. package/examples/kernel/comparison/comparison-corpus.json +337 -0
  29. package/examples/kernel/comparison/next-task-continuation.json +55 -0
  30. package/examples/kernel/corpus/operations/habanero-routing.checkpoint.kernel.yml +12 -0
  31. package/examples/kernel/corpus/operations/runner-routing.policy.kernel.yml +9 -0
  32. package/examples/kernel/corpus/product/project-home.decision.kernel.yml +11 -0
  33. package/examples/kernel/corpus/research/kernel-handoff.experiment.kernel.yml +16 -0
  34. package/examples/kernel/corpus/research/lane-drift.hypothesis.kernel.yml +11 -0
  35. package/examples/kernel/corpus/software/trace-widget.task.kernel.yml +13 -0
  36. package/examples/kernel/corpus/writing/kernel-launch.result.kernel.yml +12 -0
  37. package/llms.txt +3 -0
  38. package/package.json +4 -1
  39. package/scripts/orp-kernel-agent-pilot.py +673 -0
  40. package/scripts/orp-kernel-agent-replication.py +307 -0
  41. package/scripts/orp-kernel-benchmark.py +471 -2
  42. package/scripts/orp-kernel-canonical-continuation.py +381 -0
  43. package/scripts/orp-kernel-ci-check.py +138 -0
  44. package/scripts/orp-kernel-comparison.py +592 -0
  45. package/scripts/orp-kernel-continuation-pilot.py +384 -0
  46. package/scripts/orp-kernel-pickup.py +401 -0
  47. package/spec/v1/kernel-extension.schema.json +96 -0
  48. package/spec/v1/kernel-proposal.schema.json +115 -0
  49. package/spec/v1/kernel.schema.json +2 -1
  50. package/spec/v1/youtube-source.schema.json +151 -0
package/README.md CHANGED
@@ -23,7 +23,17 @@ verification remains independent of framing. See `modules/instruments/README.md`
23
23
  - `docs/AGENT_LOOP.md` — canonical operating loop when an agent is the primary ORP user
24
24
  - `docs/CANONICAL_CLI_BOUNDARY.md` — canonical source-of-truth boundary between CLI, Rust, and web
25
25
  - `docs/ORP_REASONING_KERNEL_V0_1.md` — draft kernel model for turning loose intent into promotable canonical artifacts
26
+ - `docs/ORP_REASONING_KERNEL_EVOLUTION.md` — stable-core, observable-pressure, explicit-evolution model for growing the kernel over time
26
27
  - `docs/ORP_REASONING_KERNEL_TECHNICAL_VALIDATION.md` — technical rationale, benchmarks, and alternatives analysis for the kernel
28
+ - `docs/ORP_REASONING_KERNEL_COMPARISON_PILOT.md` — first matched comparison between free-form, checklist, and kernel artifacts
29
+ - `docs/ORP_REASONING_KERNEL_PICKUP_PILOT.md` — first explicit pickup/handoff proxy across free-form, checklist, and kernel artifacts
30
+ - `docs/ORP_REASONING_KERNEL_AGENT_PILOT.md` — first live fresh-agent Codex recoverability simulation across free-form, checklist, and kernel artifacts
31
+ - `docs/ORP_REASONING_KERNEL_AGENT_REPLICATION.md` — `10`-repeat fresh-agent replication study for live kernel recoverability and per-field stability
32
+ - `docs/ORP_REASONING_KERNEL_CONTINUATION_PILOT.md` — first live downstream continuation smoke for kernel, checklist, and free-form artifacts
33
+ - `docs/ORP_REASONING_KERNEL_CANONICAL_CONTINUATION_PILOT.md` — harder live downstream benchmark where the agent must produce the next canonical task artifact
34
+ - `docs/ORP_REASONING_KERNEL_EVIDENCE_MATRIX.md` — honest map of what the kernel proves, only suggests, or still leaves unproven
35
+ - `docs/ORP_REASONING_KERNEL_EVALUATION_PLAN.md` — comparative experiment plan for upgrading kernel evidence beyond implementation validity
36
+ - `docs/ORP_YOUTUBE_INSPECT.md` — first-class YouTube metadata/transcript ingestion surface for agent-readable external source context
27
37
  - `docs/EXTERNAL_CONTRIBUTION_GOVERNANCE.md` — canonical local-first workflow for external OSS PR work
28
38
  - `docs/OSS_CONTRIBUTION_AGENT_LOOP.md` — agent operating rhythm for external contribution workflows
29
39
  - `templates/` — claim, verification, failure, and issue templates
@@ -42,6 +52,7 @@ verification remains independent of framing. See `modules/instruments/README.md`
42
52
  ORP should feel like one CLI with built-in abilities:
43
53
 
44
54
  - `workspace` for hosted auth, idea, feature, world, checkpoint, and worker operations
55
+ - `youtube` for public video metadata and transcript ingestion
45
56
  - `governance` for local-first repo initialization, branch safety, checkpoint commits, backup refs, readiness, repair, and cleanup
46
57
  - `discover` for profile-based GitHub scanning and opportunity selection
47
58
  - `collaborate` for repository collaboration setup and workflow execution
@@ -109,6 +120,8 @@ orp home --json
109
120
  orp about --json
110
121
  orp auth login
111
122
  orp whoami --json
123
+ orp youtube inspect https://www.youtube.com/watch?v=<video_id> --json
124
+ orp youtube inspect https://www.youtube.com/watch?v=<video_id> --save --json
112
125
  orp ideas list --json
113
126
  orp world bind --idea-id <idea-id> --project-root /abs/path --codex-session-id <session-id> --json
114
127
  orp checkpoint queue --idea-id <idea-id> --json
@@ -140,6 +153,7 @@ These surfaces are meant to help automated systems discover ORP quickly:
140
153
  - bare `orp` opens a home screen with repo/runtime status, available packs, and next commands
141
154
  - `orp home --json` returns the same landing context in machine-readable form
142
155
  - `orp auth ...`, `orp ideas ...`, `orp world ...`, `orp checkpoint ...`, `orp runner ...`, and `orp agent ...` expose the hosted workspace surface directly through ORP
156
+ - `orp youtube inspect ...` exposes public YouTube metadata and transcript retrieval through a stable ORP artifact shape for agent use
143
157
  - `orp init`, `orp status`, `orp branch start`, `orp checkpoint create`, `orp backup`, `orp ready`, `orp doctor`, and `orp cleanup` expose the local-first repo governance surface directly through ORP
144
158
  - `orp discover ...` exposes profile-based GitHub scanning as a built-in ORP ability
145
159
  - `orp collaborate ...` exposes built-in collaboration setup and workflow execution without asking users to think in terms of separate governance packs
@@ -203,6 +217,7 @@ Minimal CLI skeleton:
203
217
 
204
218
  ```bash
205
219
  orp auth login
220
+ orp youtube inspect https://www.youtube.com/watch?v=<video_id> --json
206
221
  orp ideas list --json
207
222
  orp world bind --idea-id <idea-id> --project-root /abs/path --codex-session-id <session-id> --json
208
223
  orp checkpoint queue --idea-id <idea-id> --json