@cursor/july 0.1.105 → 0.1.108

package/dist/docs/assets/guides_cloud-agents.md.DS8RIjwx.js

@@ -1,9 +0,0 @@
-import{_ as t,c as a,o as s,ag as n}from"./chunks/framework.BCISBCiQ.js";const p=JSON.parse('{"title":"Cursor cloud agents","description":"Let an agent-sdk agent hand coding work to Cursor cloud agents and steer them: mount the extension, pick the repositories it may touch, and decide what needs approval.","frontmatter":{"title":"Cursor cloud agents","description":"Let an agent-sdk agent hand coding work to Cursor cloud agents and steer them: mount the extension, pick the repositories it may touch, and decide what needs approval."},"headers":[],"relativePath":"guides/cloud-agents.md","filePath":"guides/cloud-agents.md"}'),o={name:"guides/cloud-agents.md"};function i(r,e,h,l,d,u){return s(),a("div",null,[...e[0]||(e[0]=[n(`<h1 id="cursor-cloud-agents" tabindex="-1">Cursor cloud agents <a class="header-anchor" href="#cursor-cloud-agents" aria-label="Permalink to &quot;Cursor cloud agents&quot;">​</a></h1><p>The <code>cursor-cloud-agents</code> extension lets an agent delegate coding work to Cursor cloud agents and stay in the conversation while they do it. The cloud agents work on real repository checkouts and push branches and pull requests; your agent hands off the task, reports back, and answers questions in the meantime. A Slack triage agent turns a bug report into a PR, a planner fans a refactor out across repositories, a reviewer asks a cloud agent to address its own comments.</p><p>This is different from <a href="./../reference/agent-config.html#choose-a-runtime"><code>runtime: &quot;cloud&quot;</code></a>. The runtime moves <em>this</em> agent&#39;s turns onto a cloud VM. The extension keeps the agent where it is and lets it start <em>other</em> cloud agents.</p><h2 id="mount-it" tabindex="-1">Mount it <a class="header-anchor" href="#mount-it" aria-label="Permalink to &quot;Mount it&quot;">​</a></h2><p>The extension ships with <code>@cursor/july</code>. Mount it with one file.</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:#6A737D;--shiki-dark:#6A737D;">// agent/extensions/cloud.ts</span></span>
-<span class="line"><span style="--shiki-light:#D73A49;--shiki-dark:#F97583;">import</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;"> cursorCloudAgents </span><span style="--shiki-light:#D73A49;--shiki-dark:#F97583;">from</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> &quot;@cursor/july/extensions/cursor-cloud-agents&quot;</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">;</span></span>
-<span class="line"></span>
-<span class="line"><span style="--shiki-light:#D73A49;--shiki-dark:#F97583;">export</span><span style="--shiki-light:#D73A49;--shiki-dark:#F97583;"> default</span><span style="--shiki-light:#6F42C1;--shiki-dark:#B392F0;"> cursorCloudAgents</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">({</span></span>
-<span class="line"><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">  cloud: {</span></span>
-<span class="line"><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">    repos: [{ url: </span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;">&quot;https://github.com/acme/api&quot;</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">, startingRef: </span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;">&quot;main&quot;</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;"> }],</span></span>
-<span class="line"><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">    autoCreatePR: </span><span style="--shiki-light:#005CC5;--shiki-dark:#79B8FF;">true</span><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">,</span></span>
-<span class="line"><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">  },</span></span>
-<span class="line"><span style="--shiki-light:#24292E;--shiki-dark:#E1E4E8;">});</span></span></code></pre></div><p>Launches run under the host&#39;s signed-in account (<code>agent-sdk login</code> or <code>CURSOR_API_KEY</code>) and spend its cloud-agent budget. Name the repositories the agent may target, and keep approval on when you can&#39;t.</p><p>Mount it on an agent that runs on the local runtime; that is where launches park for approval. On the cloud runtime approvals don&#39;t resolve.</p><h2 id="what-the-agent-can-do" tabindex="-1">What the agent can do <a class="header-anchor" href="#what-the-agent-can-do" aria-label="Permalink to &quot;What the agent can do&quot;">​</a></h2><p>With the extension mounted, the agent can hand off coding work and stay in the conversation while it happens:</p><ul><li><strong>Hand off a task.</strong> Launch a cloud agent on one of your repositories, or on a pull request, with a self-contained brief. It gets back an agent id and a URL to share.</li><li><strong>Check on it.</strong> See whether the agent is still running, what it produced, and which branches or pull requests it pushed.</li><li><strong>Read what was done.</strong> Pull the agent&#39;s full conversation into the workspace, quote its final report, or trace the exact edits before summarizing for a person.</li><li><strong>Steer or stop.</strong> Send a follow-up to an agent it launched; it queues behind the current run, or interrupts it on request. Cancel a run that went sideways.</li><li><strong>Keep track.</strong> List the agents launched in this conversation, or everything it has launched before.</li></ul><p>A skill teaches the model when to delegate, how to write a brief a cloud agent can act on alone, and how to follow up. Everything the extension launches stays bound to whoever launched it: an agent started by another user, or outside the extension, is out of reach. That separation is only as fine as your host&#39;s auth; a shared bearer token or <code>--allow-anonymous</code> makes every caller one principal, so use <a href="./../reference/channels.html#auth-policies">authored auth</a> when callers need their own.</p><h2 id="configure-it" tabindex="-1">Configure it <a class="header-anchor" href="#configure-it" aria-label="Permalink to &quot;Configure it&quot;">​</a></h2><table tabindex="0"><thead><tr><th>Option</th><th>What it controls</th></tr></thead><tbody><tr><td><code>cloud</code></td><td>Launch defaults, in the same shape as the agent&#39;s own <a href="./../reference/agent-config.html#cloud-options">cloud options</a>.</td></tr><tr><td><code>cloud.repos</code></td><td>The repositories the agent may hand work to, with a default branch or pull request for each. Without it, launches run with no repository and the model can&#39;t name one, so secrets in <code>cloud.envVars</code> never reach a repository you didn&#39;t list.</td></tr><tr><td><code>model</code></td><td>The model launched agents run. Omit it to use the account default.</td></tr><tr><td><code>needsApproval</code></td><td>Park launches and follow-ups for a <a href="./human-in-the-loop.html">human decision</a>. Defaults to <code>true</code> when no <code>cloud.repos</code> are configured, <code>false</code> otherwise.</td></tr><tr><td><code>apiKey</code></td><td>A Cursor API key to launch with instead of the host&#39;s own.</td></tr></tbody></table><h2 id="adjust-it" tabindex="-1">Adjust it <a class="header-anchor" href="#adjust-it" aria-label="Permalink to &quot;Adjust it&quot;">​</a></h2><p>To take a capability away, such as sending follow-ups, mount as a directory and disable that tool; see <a href="./../reference/extensions.html#adjust-a-mounted-extension">overlays</a>.</p><h2 id="what-s-next" tabindex="-1">What&#39;s next <a class="header-anchor" href="#what-s-next" aria-label="Permalink to &quot;What&#39;s next&quot;">​</a></h2><p>Continue with these pages:</p><ul><li><a href="./../reference/extensions.html">Extensions</a>: how mounts, namespaces, and overrides work</li><li><a href="./human-in-the-loop.html">Human-in-the-loop approvals</a>: where parked launches get approved</li></ul>`,19)])])}const g=t(o,[["render",i]]);export{p as __pageData,g as default};

package/dist/docs/assets/index.md.BW_6tOgR.js

@@ -1,5 +0,0 @@
-import{_ as t,c as a,o as s,ag as r}from"./chunks/framework.BCISBCiQ.js";const g=JSON.parse('{"title":"Agent SDK documentation","description":"Build and run Cursor agents with tools, approvals, channels, and evals.","frontmatter":{"title":"Agent SDK documentation","description":"Build and run Cursor agents with tools, approvals, channels, and evals."},"headers":[],"relativePath":"index.md","filePath":"README.md"}'),i={name:"index.md"};function n(o,e,l,h,d,c){return s(),a("div",null,[...e[0]||(e[0]=[r(`<h1 id="agent-sdk-documentation" tabindex="-1">Agent SDK documentation <a class="header-anchor" href="#agent-sdk-documentation" aria-label="Permalink to &quot;Agent SDK documentation&quot;">​</a></h1><p>Use the Agent SDK to define Cursor agents in TypeScript and Markdown. See <a href="./reference/project-layout.html">Project layout</a> for the directory structure.</p><p>Use Node 22.13 or newer. Bun isn&#39;t supported.</p><p>Create a project:</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;">npx</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> @cursor/july</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> init</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> ./my-agent</span></span>
-<span class="line"><span style="--shiki-light:#005CC5;--shiki-dark:#79B8FF;">cd</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> my-agent</span></span>
-<span class="line"><span style="--shiki-light:#6F42C1;--shiki-dark:#B392F0;">agent-sdk</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> dev</span></span></code></pre></div><p>If <code>agent-sdk</code> isn&#39;t on <code>PATH</code>, use <code>npx @cursor/july &lt;command&gt;</code>.</p><p>Open the docs locally:</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;">npx</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> @cursor/july</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> docs</span></span></code></pre></div><h2 id="where-to-start" tabindex="-1">Where to start <a class="header-anchor" href="#where-to-start" aria-label="Permalink to &quot;Where to start&quot;">​</a></h2><table tabindex="0"><thead><tr><th>You are...</th><th>Start with</th></tr></thead><tbody><tr><td>New to the Agent SDK</td><td><a href="./quickstart.html">Quickstart</a> (PR reviewer), then <a href="./concepts.html">Concepts</a></td></tr><tr><td>Building a new agent with Cursor</td><td><a href="./scaffolding-agents.html">Scaffold an agent with Cursor</a></td></tr><tr><td>Turning a Cursor Automation into a project</td><td><a href="./guides/convert-automation.html">Convert a Cursor Automation</a></td></tr><tr><td>Wiring an agent to Slack</td><td><a href="./guides/slack.html">Slack guide</a></td></tr><tr><td>Starting from a packaged template</td><td><a href="./templates/demo.html">Demo</a>, <a href="./templates/grokbot-agents.html">Grok Bot agents</a>, <a href="./templates/code-wiki.html">Code wiki</a>, <a href="./templates/agents-md.html">Living AGENTS.md</a>, <a href="./templates/security-reviewer.html">Security reviewer</a>, <a href="./templates/security-help.html">Security help</a>, <a href="./templates/triage.html">Triage</a>, or <a href="./templates/agentic-owners.html">Agentic Owners</a></td></tr><tr><td>Wiring an agent to GitHub webhooks</td><td><a href="./guides/github.html">GitHub guide</a></td></tr><tr><td>Driving PRs from a cloud VM</td><td><a href="./templates/pr-autofixer.html">PR autofixer template</a></td></tr><tr><td>Handing coding work to Cursor cloud agents</td><td><a href="./guides/cloud-agents.html">Cursor cloud agents</a></td></tr><tr><td>Talking to your Grok Bot agents</td><td><a href="./guides/grokbot-agents.html">Cursor Grok Bot agents</a></td></tr><tr><td>Driving an agent from Linear (or another tracker)</td><td><a href="./guides/webhooks.html#example-linear-as-the-control-plane">Webhooks guide: Linear example</a></td></tr><tr><td>Making an existing agent measurably better</td><td><a href="./evals.html">Evals</a>, then <a href="./hillclimbing.html">Hillclimbing</a></td></tr><tr><td>Comparing variants on live traffic</td><td><a href="./ab.html">Live A/B metrics</a></td></tr><tr><td>Deploying with Cursor or on your own infrastructure</td><td><a href="./deployment.html">Deployment</a></td></tr><tr><td>Debugging something that misbehaves</td><td><a href="./troubleshooting.html">Fix common agent problems</a></td></tr></tbody></table><h2 id="documentation" tabindex="-1">Documentation <a class="header-anchor" href="#documentation" aria-label="Permalink to &quot;Documentation&quot;">​</a></h2><p><strong>Core</strong></p><ul><li><a href="./quickstart.html">Quickstart</a>: build a PR reviewer that classifies changes by complexity and handles GitHub webhook events.</li><li><a href="./scaffolding-agents.html">Scaffold an agent with Cursor</a>: use the bundled skill for a guided build.</li><li><a href="./guides/convert-automation.html">Convert a Cursor Automation</a>: export a dashboard Automation into an Agent SDK project.</li><li><a href="./concepts.html">Concepts</a>: agent discovery, sessions, channels, runtimes, and observability.</li></ul><p><strong>Templates</strong></p><ul><li><a href="./templates/demo.html">Record a walkthrough from a collected PR</a>: host collects the PR, the model records, then comments.</li><li><a href="./templates/grokbot-agents.html">Talk to your Grok Bot agents</a>: name the bots, then consult them from this agent&#39;s own conversation.</li><li><a href="./templates/code-wiki.html">Keep a code wiki current after merge</a>: update in-repo <code>wiki/</code> pages when a pull request merges.</li><li><a href="./templates/agents-md.html">Keep AGENTS.md current from last week&#39;s work</a>: pull rules out of merged PRs and Slack, then open a follow-up PR.</li><li><a href="./templates/security-reviewer.html">Security reviewer</a>: review pull requests for exploitable bugs and post one comment.</li><li><a href="./templates/security-help.html">Answer security questions in Slack</a>: answer from your skills, or take a first look at a pasted PR.</li><li><a href="./templates/triage.html">Triage Linear or Jira issues in place</a>: classify existing tickets and comment on them.</li><li><a href="./templates/agentic-owners.html">Review pull requests with owners policies</a>: request owners and approve changes allowed by repository policy.</li><li><a href="./templates/pr-autofixer.html">Fix pull requests on a Cursor cloud VM</a></li></ul><p><strong>Self-improving Agents</strong></p><ul><li><a href="./building-with-agents.html">Building agents with agents</a>: use a coding agent to scaffold, run, and iterate on your agent.</li><li><a href="./evals.html">Evals</a>: author <code>defineEval</code> cases, pick fixtures, and use evals as regression checks.</li><li><a href="./ab.html">Live A/B metrics</a>: assign sticky variants and compare cumulative metrics on live sessions.</li><li><a href="./storage.html">Storage</a>: point durable storage at a backend you own with <code>defineStorage</code>.</li><li><a href="./hillclimbing.html">Hillclimbing</a>: measure and improve an agent iteratively.</li></ul><p><strong>Guides</strong></p><ul><li><a href="./guides/webhooks.html">Webhooks and custom channels</a>: give the agent its own HTTP surface.</li><li><a href="./guides/github.html">GitHub</a>: trigger the agent from pull requests, CI, and comments.</li><li><a href="./guides/slack.html">Slack</a>: put the agent in Slack over Socket Mode.</li><li><a href="./guides/human-in-the-loop.html">Human-in-the-loop approvals</a>: park a tool call until a person signs off.</li><li><a href="./guides/mcp-oauth.html">Host MCP OAuth</a>: authorize <code>oauth: true</code> connections, store tokens locally, and persist them on hosted deployments (<code>--store</code>).</li><li><a href="./guides/agent-to-agent.html">Agent-to-agent</a>: every agent is an MCP server; agents can delegate to each other.</li><li><a href="./guides/cloud-agents.html">Cursor cloud agents</a>: let the agent launch and steer cloud agents through the <code>cursor-cloud-agents</code> extension.</li><li><a href="./guides/grokbot-agents.html">Cursor Grok Bot agents</a>: let the agent talk to Grok Bot agents through the <code>cursor-grokbot-agents</code> extension.</li><li><a href="./guides/opentelemetry.html">OpenTelemetry</a>: push session, turn, and tool traces to an OTLP collector you run.</li></ul><p><strong>Operating</strong></p><ul><li><a href="./deployment.html">Deployment</a>: Cursor-managed hosting, self-hosting, auth, state, and operations.</li><li><a href="./troubleshooting.html">Fix common agent problems</a>: diagnose common failures by symptom.</li></ul><p><strong>Reference</strong></p><ul><li><a href="./reference/project-layout.html">Project layout</a>: the full folder structure.</li><li><a href="./reference/agent-config.html">Agent config</a> · <a href="./reference/instructions.html">Instructions</a> · <a href="./reference/tools.html">Tools</a> · <a href="./reference/prompt.html"><code>prompt</code></a> · <a href="./reference/skills.html">Skills</a> · <a href="./reference/connections.html">MCP connections</a> · <a href="./reference/subagents.html">Subagents</a> · <a href="./reference/extensions.html">Extensions</a></li><li><a href="./reference/channels.html">Channels</a> · <a href="./reference/schedules.html">Schedules and reminders</a> · <a href="./reference/hooks.html">Hooks</a> · <a href="./reference/artifacts.html">Artifacts</a> · <a href="./reference/result.html">Turn result</a> · <a href="./reference/sessions.html">Sessions and streaming</a> · <a href="./reference/playground.html">Playground</a></li><li><a href="./reference/cli.html">CLI</a> · <a href="./reference/http-api.html">HTTP API</a></li></ul><h2 id="run-the-cli" tabindex="-1">Run the CLI <a class="header-anchor" href="#run-the-cli" aria-label="Permalink to &quot;Run the CLI&quot;">​</a></h2><p>Docs use <code>agent-sdk &lt;command&gt;</code>. If it isn&#39;t on <code>PATH</code>, use <code>npx @cursor/july &lt;command&gt;</code>.</p><h2 id="credentials" tabindex="-1">Credentials <a class="header-anchor" href="#credentials" aria-label="Permalink to &quot;Credentials&quot;">​</a></h2><p>Sign in to Cursor or set <code>CURSOR_API_KEY</code>:</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;"> login</span></span>
-<span class="line"><span style="--shiki-light:#6A737D;--shiki-dark:#6A737D;"># or: export CURSOR_API_KEY=key_...</span></span>
-<span class="line"><span style="--shiki-light:#6F42C1;--shiki-dark:#B392F0;">agent-sdk</span><span style="--shiki-light:#032F62;--shiki-dark:#9ECBFF;"> whoami</span></span></code></pre></div><p>Confirm <code>agent-sdk whoami</code> shows the expected account.</p><h2 id="related-documentation" tabindex="-1">Related documentation <a class="header-anchor" href="#related-documentation" aria-label="Permalink to &quot;Related documentation&quot;">​</a></h2><ul><li>Package reference: <a href="./../README.html"><code>README.md</code></a></li><li>Coding-agent workflows: <a href="./skills/"><code>skills/</code></a></li></ul>`,31)])])}const p=t(i,[["render",n]]);export{g as __pageData,p as default};

package/dist/docs/assets/skills_framework-map.md.haibFyoB.js

@@ -1 +0,0 @@
-import{_ as t,c as d,o,ag as r}from"./chunks/framework.BCISBCiQ.js";const g=JSON.parse('{"title":"Agent SDK framework map","description":"Orient on Agent SDK layout, local vs cloud runtime, sessions, and invariants (Node not Bun, root evals/, typecheck after tsx). Use when creating, editing, or running a project with agent/agent.ts.","frontmatter":{"name":"agentsdk-framework-map","description":"Orient on Agent SDK layout, local vs cloud runtime, sessions, and invariants (Node not Bun, root evals/, typecheck after tsx). Use when creating, editing, or running a project with agent/agent.ts."},"headers":[],"relativePath":"../skills/framework-map/SKILL.md","filePath":"../skills/framework-map/SKILL.md"}'),s={name:"../skills/framework-map/SKILL.md"};function c(a,e,n,l,i,u){return o(),d("div",null,[...e[0]||(e[0]=[r('<h1 id="agent-sdk-framework-map" tabindex="-1">Agent SDK framework map <a class="header-anchor" href="#agent-sdk-framework-map" aria-label="Permalink to &quot;Agent SDK framework map&quot;">​</a></h1><p><code>@cursor/july</code> discovers files under <code>agent/</code> and serves the agent over HTTP, Slack, and GitHub. Markdown is prose. TypeScript is typed behavior. Ground truth: package <code>README.md</code> and <code>AGENTS.md</code>.</p><p>CLI is <code>agent-sdk</code> (Node, never Bun).</p><p>Public docs: <code>node_modules/@cursor/july/dist/docs/llms.txt</code> or <code>/docs/llms.txt</code> on a running host.</p><h2 id="invariants" tabindex="-1">Invariants <a class="header-anchor" href="#invariants" aria-label="Permalink to &quot;Invariants&quot;">​</a></h2><ol><li><strong>Node 22.13+, never Bun.</strong> Bun corrupts harness tool-result streams (<code>NGHTTP2_FRAME_SIZE_ERROR</code>).</li><li><strong>Evals live at project-root <code>evals/</code>.</strong> <code>agent/evals/</code> is ignored.</li><li><strong>tsx does not typecheck.</strong> Tool <code>execute</code> must return JSON-shaped values: object literals or <code>type</code> aliases, not <code>interface</code> types.</li><li><strong>Nested git checkouts.</strong> Discovery sets <code>local.cwd</code> to a per-project cache directory under <code>~/.cache</code>. Point cwd at a checkout only when the agent must inherit that tree.</li><li><strong>Attached MCP is nameless</strong> until <code>advertiseTools: true</code>.</li><li><strong>Model turns need <code>CURSOR_API_KEY</code>.</strong> <code>validate</code>, <code>info</code>, <code>call</code>, and <code>serve</code> bring-up do not.</li></ol><h2 id="folder-structure" tabindex="-1">Folder structure <a class="header-anchor" href="#folder-structure" aria-label="Permalink to &quot;Folder structure&quot;">​</a></h2><p>Path is identity. Full list: README &quot;Folder structure&quot;.</p><table tabindex="0"><thead><tr><th>Path</th><th>Role</th></tr></thead><tbody><tr><td><code>agent/agent.ts</code></td><td><code>defineAgent({ model?, runtime?, cloud?, local? })</code></td></tr><tr><td><code>agent/instructions.md</code></td><td>Always-on system prompt (required)</td></tr><tr><td><code>agent/tools/&lt;name&gt;.ts</code></td><td>One tool. <code>execution: &quot;server&quot;</code> or <code>&quot;agent&quot;</code></td></tr><tr><td><code>agent/skills/*</code></td><td>On-demand procedures</td></tr><tr><td><code>agent/mcp-connections/&lt;name&gt;.ts</code></td><td>MCP. Never name an account file <code>cursor.ts</code>. <code>advertiseTools: true</code> for named local tools</td></tr><tr><td><code>agent/host-connections/&lt;name&gt;.ts</code></td><td>Privileged MCP for <code>ctx.host.mcp</code> / <code>mcp oauth</code></td></tr><tr><td><code>agent/subagents/&lt;id&gt;/</code></td><td>Child agent (<code>description</code> required)</td></tr><tr><td><code>agent/extensions/&lt;ns&gt;.ts</code></td><td>Mounted extension or Cursor plugin (<code>cursorPlugin</code>); contributions are <code>&lt;ns&gt;__&lt;name&gt;</code></td></tr><tr><td><code>agent/channels/*.ts</code></td><td>Slack / GitHub / custom HTTP</td></tr><tr><td><code>agent/hooks/*.ts</code></td><td>Observe-only</td></tr><tr><td><code>agent/artifacts.ts</code></td><td>Durable tagged outputs (<code>defineArtifacts</code>)</td></tr><tr><td><code>agent/result.ts</code></td><td>Host <code>commit</code> on the final assistant text (<code>defineResult</code>)</td></tr><tr><td><code>agent/ab.ts</code> or <code>agent/ab/*.ts</code></td><td>Live A/B (<code>defineAB</code>)</td></tr><tr><td><code>agent/otel.ts</code></td><td>OpenTelemetry (<code>defineOtel</code>)</td></tr><tr><td><code>agent/schedules/*</code></td><td>Cron. Never auto-fire under <code>--dev</code></td></tr><tr><td><code>agent/sandbox/workspace/</code></td><td>Session seed files (local only)</td></tr><tr><td><code>agent/lib/</code></td><td>Import-only. Never discovered</td></tr><tr><td><code>evals/**/*.eval.ts</code></td><td>Case id is the path under <code>evals/</code></td></tr></tbody></table><h2 id="local-vs-cloud" tabindex="-1">Local vs cloud <a class="header-anchor" href="#local-vs-cloud" aria-label="Permalink to &quot;Local vs cloud&quot;">​</a></h2><p><code>runtime: &quot;local&quot;</code> (default) runs on the serve host. <code>runtime: &quot;cloud&quot;</code> needs <code>cloud: { repos: [...] }</code>.</p><table tabindex="0"><thead><tr><th>Capability</th><th>local</th><th>cloud</th></tr></thead><tbody><tr><td>Server tools</td><td>yes</td><td>yes on managed hosting; self-hosted needs <code>--public-url</code></td></tr><tr><td>Tool approvals</td><td>yes</td><td>no</td></tr><tr><td>Agent tools / skills</td><td>yes</td><td>yes</td></tr><tr><td>sandbox seeds</td><td>yes</td><td>no</td></tr><tr><td>Checkout</td><td>you arrange it</td><td>the VM carries it</td></tr></tbody></table><p>Use cloud when the job needs a checkout at scale. <code>validate</code> warns when cloud is combined with local-only capabilities.</p><h2 id="sessions" tabindex="-1">Sessions <a class="header-anchor" href="#sessions" aria-label="Permalink to &quot;Sessions&quot;">​</a></h2><ul><li><strong>continuationToken</strong> continues a conversation. HTTP follow-ups rotate it. Stale tokens return <code>409</code>.</li><li><strong>sessionId</strong> is the inspect handle (<code>GET /v1/session/:id/stream?startIndex=N</code>).</li></ul><p>A follow-up to a busy HTTP/MCP session interrupts the in-flight turn. Slack coalesces. Routes: <code>docs/reference/http-api.md</code>. Session files live under the project state directory (<code>--state-root</code>).</p><h2 id="where-logic-belongs" tabindex="-1">Where logic belongs <a class="header-anchor" href="#where-logic-belongs" aria-label="Permalink to &quot;Where logic belongs&quot;">​</a></h2><p>Code: side-effect gates, write dedupe, auth, evidence seeding, hard API caps. Model: formatting, summarizing, classification, replies. Budget: <code>skills/create-agent/SKILL.md</code>.</p><p>Loop: <code>validate</code> / <code>info</code> / <code>call</code> / <code>run</code> / <code>eval</code> / <code>serve</code>. Serve only this agent&#39;s directory.</p><table tabindex="0"><thead><tr><th>Task</th><th>Skill</th></tr></thead><tbody><tr><td>Scaffold</td><td><code>skills/create-agent/SKILL.md</code></td></tr><tr><td>Evals</td><td><code>skills/evals/SKILL.md</code></td></tr><tr><td>Live A/B</td><td><code>skills/ab/SKILL.md</code></td></tr><tr><td>OpenTelemetry</td><td><code>skills/otel/SKILL.md</code></td></tr><tr><td>GitHub</td><td><code>skills/github/SKILL.md</code></td></tr><tr><td>Slack</td><td><code>skills/setup-slack/SKILL.md</code></td></tr><tr><td>Host MCP OAuth</td><td><code>skills/mcp-auth/SKILL.md</code></td></tr><tr><td>Local triage</td><td><code>skills/debug/SKILL.md</code></td></tr><tr><td>Measured improvement</td><td><code>skills/hillclimb/SKILL.md</code></td></tr></tbody></table>',20)])])}const p=t(s,[["render",c]]);export{g as __pageData,p as default};

package/dist/docs/assets/troubleshooting.md.Cus_YZga.js

@@ -1 +0,0 @@
-import{_ as t,c as o,o as a,ag as d}from"./chunks/framework.BCISBCiQ.js";const p=JSON.parse('{"title":"Fix common agent problems","description":"Match what you see to a cause, then read a session trace when you need more detail.","frontmatter":{"title":"Fix common agent problems","description":"Match what you see to a cause, then read a session trace when you need more detail."},"headers":[],"relativePath":"troubleshooting.md","filePath":"troubleshooting.md"}'),r={name:"troubleshooting.md"};function s(n,e,c,i,l,h){return a(),o("div",null,[...e[0]||(e[0]=[d('<h1 id="fix-common-agent-problems" tabindex="-1">Fix common agent problems <a class="header-anchor" href="#fix-common-agent-problems" aria-label="Permalink to &quot;Fix common agent problems&quot;">​</a></h1><p>Start with four checks, in order:</p><ol><li>Project discovery: <code>agent-sdk validate --dir .</code></li><li>Whether the serve process is running</li><li>What the playground or HTTP API shows</li><li>The session event stream (trace)</li></ol><p>Match your symptom below. Keep the commands as <code>agent-sdk</code>. If it is not on <code>PATH</code>, use <code>npx @cursor/july</code>.</p><h2 id="what-if-serve-or-the-playground-looks-wrong" tabindex="-1">What if serve or the playground looks wrong? <a class="header-anchor" href="#what-if-serve-or-the-playground-looks-wrong" aria-label="Permalink to &quot;What if serve or the playground looks wrong?&quot;">​</a></h2><table tabindex="0"><thead><tr><th>What you see</th><th>What to do</th></tr></thead><tbody><tr><td><code>serve</code> won&#39;t start</td><td>Run <code>agent-sdk validate --dir .</code> and fix the reported errors.</td></tr><tr><td>Playground is blank or says there are no agents</td><td>The UI needs a running <code>serve</code> process. Building the playground assets alone is not enough.</td></tr><tr><td>The playground UI looks stale</td><td><code>serve --dev</code> prints a playground URL. Open that URL. Agent-file edits still need a restart (press Enter on the TTY).</td></tr><tr><td>Sessions exist but the playground list is empty</td><td>The list shows sessions for the authenticated caller. In <code>--dev</code> on loopback the list is wider. Otherwise open <code>/&lt;slug&gt;/playground?sessionId=ses_…</code>.</td></tr><tr><td>Port 3000 is already in use</td><td>For the default serve port, the CLI tries the next free port and prints a notice. Pass <code>--port</code> to pick one, or <code>--port 0</code> for any free port. Stop leftover playground or webhook-forwarder processes if you need the original port.</td></tr></tbody></table><h2 id="what-if-a-model-turn-goes-wrong" tabindex="-1">What if a model turn goes wrong? <a class="header-anchor" href="#what-if-a-model-turn-goes-wrong" aria-label="Permalink to &quot;What if a model turn goes wrong?&quot;">​</a></h2><table tabindex="0"><thead><tr><th>What you see</th><th>What to do</th></tr></thead><tbody><tr><td>Built-in file reads and greps fail; the turn retries for a long time</td><td>Run under Node 22.13+ (or <code>tsx</code>), never Bun. Look for <code>NGHTTP2_FRAME_SIZE_ERROR</code> in logs.</td></tr><tr><td>The turn fails immediately with an API-key error</td><td>Sign in with <code>agent-sdk login</code>, or set <code>CURSOR_API_KEY</code>. Discovery, <code>info</code>, <code>call</code>, and serve bring-up work without a key; model turns need one.</td></tr><tr><td>Replies quote rules or <code>AGENTS.md</code> from outside your agent project</td><td>The session workspace inherited parent-folder config. Nested git checkouts default <code>local.cwd</code> to a per-project cache directory under <code>~/.cache</code>. Point <code>defineAgent({ local: { cwd } })</code> at a checkout only when the agent should inherit that tree, or set <code>--state-root</code> to a clean directory (for example under <code>/tmp</code>).</td></tr><tr><td>Yellow box shows Datadog/Linear tools, but the model lists <code>GetDynamicTools</code> / IDE <code>cursor</code> tools and never calls them</td><td>Attached MCP sits behind harness meta-tools, or the harness cwd is still inside another checkout. Set <code>advertiseTools: true</code> for named tools on local turns. Check <code>GET /v1/info</code> <code>local.cwd</code> and <code>connections[].advertiseTools</code>.</td></tr><tr><td>Server tools, skills, or workspace seed files never appear</td><td>Server tools and sandbox seeds apply on the local runtime (cloud server tools need <code>--public-url</code> / <code>--cloud-tools-url</code>). Skills reach cloud through the Agent Store when hosting or a personal <code>CURSOR_API_KEY</code> is available; otherwise only skills already in the cloud repo. <code>validate</code> warns when this combination is present.</td></tr><tr><td><code>validate</code> and <code>run</code> succeed, but typecheck fails in CI</td><td>The CLI runs TypeScript with type-stripping only. Keep tool <code>execute</code> return types as object literals or <code>type</code> aliases, not <code>interface</code> types.</td></tr><tr><td>Login works, but turns are rejected when using custom API hosts</td><td>Point login and model traffic at the same host (<code>CURSOR_API_BASE_URL</code> and <code>CURSOR_BACKEND_URL</code>). A key from one host is rejected by the other.</td></tr></tbody></table><h2 id="what-if-the-http-api-returns-an-error" tabindex="-1">What if the HTTP API returns an error? <a class="header-anchor" href="#what-if-the-http-api-returns-an-error" aria-label="Permalink to &quot;What if the HTTP API returns an error?&quot;">​</a></h2><table tabindex="0"><thead><tr><th>What you see</th><th>What to do</th></tr></thead><tbody><tr><td><code>409</code> on a follow-up message</td><td>Refresh the <code>continuationToken</code> or confirm the session is a chat session. Task sessions do not accept follow-ups.</td></tr><tr><td><code>409 session_busy</code> on <code>call --session</code></td><td>The tool is a write (or declares no effect) and a model turn is running. Wait for the turn to finish, omit <code>--session</code> for a one-off call, or declare <code>effect: &quot;read&quot;</code> on a tool that only reads so it runs alongside turns.</td></tr><tr><td><code>403</code> on stream or follow-up</td><td>Use the same auth identity that created the session. Off localhost, pass <code>--bearer-token</code> and send it on every request.</td></tr><tr><td>Works on localhost; blocked through a tunnel or LAN</td><td>Default auth allows only direct loopback callers. Share the host with <code>--bearer-token &lt;secret&gt;</code> (or authored <code>bearerAuth</code>). Use <code>--allow-anonymous</code> only on a trusted private network.</td></tr><tr><td>A channel route fails to compile with a schema type error</td><td><code>GET</code> routes need a Zod <code>querySchema</code>. <code>POST</code> / <code>PUT</code> / <code>PATCH</code> need a Zod <code>bodySchema</code>. Use <code>z.object({})</code> or <code>z.unknown()</code> for open shapes.</td></tr></tbody></table><h2 id="what-if-github-webhooks-misbehave" tabindex="-1">What if GitHub webhooks misbehave? <a class="header-anchor" href="#what-if-github-webhooks-misbehave" aria-label="Permalink to &quot;What if GitHub webhooks misbehave?&quot;">​</a></h2><table tabindex="0"><thead><tr><th>What you see</th><th>What to do</th></tr></thead><tbody><tr><td><code>github forward</code> returns 401 on every delivery, but the hook was created</td><td>Clear <code>GITHUB_TOKEN</code> and <code>GH_TOKEN</code> for that command. The forwarder uses your <code>gh</code> CLI login: <code>GITHUB_TOKEN= GH_TOKEN= agent-sdk github forward …</code></td></tr><tr><td><code>Hook already exists</code> when starting a forwarder</td><td>GitHub allows one forwarder per repo. Run a single <code>github forward --dir &lt;parent&gt;</code> and stop stale forwarders.</td></tr><tr><td>Deliveries rejected outside <code>--dev</code></td><td>Set <code>GITHUB_WEBHOOK_SECRET</code> on the server and on the signer. Without a secret, the channel stays loopback-only.</td></tr><tr><td>You lack repo admin and can&#39;t forward</td><td>Use <code>agent-sdk github replay &lt;pr-url&gt;</code>. It needs pull access only and posts signed test payloads.</td></tr></tbody></table><h2 id="what-if-slack-stays-quiet" tabindex="-1">What if Slack stays quiet? <a class="header-anchor" href="#what-if-slack-stays-quiet" aria-label="Permalink to &quot;What if Slack stays quiet?&quot;">​</a></h2><table tabindex="0"><thead><tr><th>What you see</th><th>What to do</th></tr></thead><tbody><tr><td>Logs show <code>channel idle … missing credentials</code></td><td>Expected when tokens are missing. Run <code>agent-sdk slack create --dir &lt;agent&gt;</code> to provision the app and write the tokens, or <code>agent-sdk slack init --manual --dir &lt;agent&gt;</code> and paste the manifests at api.slack.com. Then set <code>&lt;PREFIX&gt;_SLACK_BOT_TOKEN</code> and <code>&lt;PREFIX&gt;_SLACK_APP_TOKEN</code> per agent and run <code>agent-sdk slack doctor --prefix &lt;PREFIX&gt;</code>.</td></tr><tr><td><code>slack create</code> reports the app needs admin approval</td><td>Open Slack&#39;s <strong>Request approval</strong> page (the CLI prints the link; the same URL is <strong>Send a reminder</strong> after you submit). Managed install does not file the request. Keep the CLI running, then click <strong>Retry</strong> in the dashboard after an admin approves.</td></tr><tr><td>The bot ignores ordinary channel posts</td><td>Default engagement is mentions and DMs only. Enable <code>engagement.channelPosts</code> with an allowlist, and subscribe the app to <code>message.channels</code> / <code>message.groups</code>.</td></tr><tr><td>Approve / Deny buttons do nothing</td><td>Channels that post approval cards need <code>toolApprovals: true</code>. Recreate the app with <code>slack create</code> if interactivity is off.</td></tr></tbody></table><h2 id="what-if-host-mcp-oauth-fails" tabindex="-1">What if host MCP OAuth fails? <a class="header-anchor" href="#what-if-host-mcp-oauth-fails" aria-label="Permalink to &quot;What if host MCP OAuth fails?&quot;">​</a></h2><table tabindex="0"><thead><tr><th>What you see</th><th>What to do</th></tr></thead><tbody><tr><td><code>must be defineConnection({ url, oauth: true })</code></td><td>The connection file needs <code>oauth: true</code>, or you passed the wrong connection name to <code>agent-sdk mcp oauth</code>.</td></tr><tr><td>Local auth works; hosted calls unauthorized</td><td>Finish Connect on the hosted engine, or run <code>agent-sdk mcp oauth &lt;name&gt; --store</code>. Confirm names with <code>agent-sdk secrets list &lt;slug&gt;</code>, then redeploy.</td></tr><tr><td>Model asks for <code>mcp_auth</code> or IDE MCP for a connector it already has</td><td>Attached MCP is behind meta-tools. Set <code>advertiseTools: true</code> for named tools on local turns, or call it from a host tool via <code>ctx.host.mcp</code>.</td></tr></tbody></table><p>See <a href="./guides/mcp-oauth.html">Host MCP OAuth</a> and <a href="./skills/mcp-auth.html"><code>skills/mcp-auth/SKILL.md</code></a>.</p><h2 id="what-if-a-secret-showed-up-in-a-terminal-transcript" tabindex="-1">What if a secret showed up in a terminal transcript? <a class="header-anchor" href="#what-if-a-secret-showed-up-in-a-terminal-transcript" aria-label="Permalink to &quot;What if a secret showed up in a terminal transcript?&quot;">​</a></h2><table tabindex="0"><thead><tr><th>What you see</th><th>What to do</th></tr></thead><tbody><tr><td><code>secrets set … NAME=VALUE</code> in an agent-captured terminal or shell history</td><td>Rotate the secret at the provider. Set it again with names only: <code>agent-sdk secrets set &lt;slug&gt; NAME</code> (hidden prompt) or pipe/redirect the value. <code>NAME=VALUE</code> requires <code>--from-argv</code> and still leaks into argv.</td></tr><tr><td>Alias token printed during first deploy or <code>rotate-token</code></td><td>Treat it as exposed if the transcript left your machine. Run <code>agent-sdk rotate-token &lt;slug&gt;</code>, store the new token outside agent transcripts, and update callers.</td></tr><tr><td>Someone verified a secret with <code>echo</code> / <code>printenv</code></td><td>Rotate it. Confirm presence with <code>agent-sdk secrets list &lt;slug&gt;</code> (names only), then redeploy and test the feature.</td></tr></tbody></table><h2 id="what-if-schedules-reminders-or-approvals-stall" tabindex="-1">What if schedules, reminders, or approvals stall? <a class="header-anchor" href="#what-if-schedules-reminders-or-approvals-stall" aria-label="Permalink to &quot;What if schedules, reminders, or approvals stall?&quot;">​</a></h2><table tabindex="0"><thead><tr><th>What you see</th><th>What to do</th></tr></thead><tbody><tr><td>A schedule or reminder never fires under <code>--dev</code></td><td>Dev mode does not auto-fire. Trigger with <code>POST /&lt;slug&gt;/v1/dev/schedules/&lt;id&gt;</code> or <code>POST /&lt;slug&gt;/v1/dev/reminders/&lt;id&gt;</code> (list reminders at <code>GET /v1/dev/reminders</code>).</td></tr><tr><td>A pending tool approval disappeared after restart</td><td>Parked approvals do not survive host restart. They resolve as interrupted. Run the turn again.</td></tr><tr><td>A reminder is disarmed after restart (<code>handler_lost_on_restart</code>)</td><td>Handler-form reminders live in memory. Re-arm them from the code that created them, or use prompt-form reminders.</td></tr></tbody></table><h2 id="how-do-i-read-a-session-trace" tabindex="-1">How do I read a session trace? <a class="header-anchor" href="#how-do-i-read-a-session-trace" aria-label="Permalink to &quot;How do I read a session trace?&quot;">​</a></h2><p>Look at <code>actions.requested</code> / <code>action.result</code> pairs for the tool trajectory. Count calls by tool name before blaming latency. Separate host-side work (channel <code>callTool</code>, preparation) from tools the model chose.</p><p><code>turn.failed</code> with <code>&quot;turn interrupted&quot;</code> means a follow-up or stop ended the turn on purpose.</p><p>If the model reads outside the session workspace, the prepared files don&#39;t match what the instructions expect. Fix the layout. See <a href="./hillclimbing.html">Hillclimbing</a>.</p><p><code>agent-sdk trajectory --events &lt;file&gt;</code> summarizes any saved NDJSON stream. The playground <strong>Open trace</strong> control does the same visually.</p><h2 id="what-s-next" tabindex="-1">What&#39;s next <a class="header-anchor" href="#what-s-next" aria-label="Permalink to &quot;What&#39;s next&quot;">​</a></h2><ul><li><a href="./concepts.html">Concepts</a>: the model behind these symptoms</li><li><a href="./hillclimbing.html">Hillclimbing</a>: when the agent runs but underperforms</li><li><a href="./deployment.html">Deployment</a>: auth and state on shared hosts</li></ul>',28)])])}const m=t(r,[["render",s]]);export{p as __pageData,m as default};