@cursor/july 0.1.95 → 0.1.97

package/skills/evals/SKILL.md

@@ -1,129 +1,58 @@
 ---
 name: agentsdk-evals
 description: >-
-  Author and run Agent SDK defineEval cases (single- or multi-datapoint).
-  Use when writing, fixing, seeding, or hillclimbing evals; AskQuestion
-  whether to generate samples or upload data. Materialize API-backed
-  fixtures first. Live A/B metrics: defineAB (skills/ab), not defineEval.
+  Author and run defineEval cases. Use when writing, fixing, seeding,
+  or hillclimbing evals. Ask whether to generate samples or take
+  uploaded data. Materialize API-backed fixtures first. Live A/B is
+  defineAB (skills/ab), not defineEval.
 ---
 
 # Agent SDK evals
 
-Public docs index: `node_modules/@cursor/july/dist/docs/llms.txt`. A running
-host serves the same index at `/docs/llms.txt`.
+Fixed input, model turn, gates on the trajectory. Files live at
+project-root `evals/**/*.eval.ts`. `agent/evals/` is ignored.
 
-Evals are the ratchet that makes iteration trustworthy: a fixed input, a
-model turn, and gates over the recorded trajectory. They live at the
-**project root** `evals/**/*.eval.ts` (never `agent/evals/` — that slot is
-ignored).
-
-For **live A/B metrics** while the agent serves traffic, use `defineAB`
-under `agent/ab` (`packages/agent-serve/skills/ab/SKILL.md`). That plug-in
-enrolls via sticky `split` and calls `onSample`; it is not a test runner
-and does not replace `defineEval` gates.
-
-**Identity**
-
-- File id = path under `evals/` without `.eval.ts`
-  (`evals/weather.eval.ts` → `weather`).
-- Single datapoint (`test`): case id = file id
-  (`evals/weather/nyc.eval.ts` → `weather/nyc`).
-- Multiple datapoints (`cases`): case id = `<fileId>/<case.id>`
-  (`evals/weather.eval.ts` + `{ id: "nyc" }` → `weather/nyc`).
+Live traffic variants: `skills/ab/SKILL.md`. That is not a test
+runner.
 
 ```bash
-agent-sdk eval --dir . --list              # all datapoints
-agent-sdk eval --dir . --json              # run all
-agent-sdk eval --dir . weather/nyc         # one datapoint
-agent-sdk eval --dir . weather             # every datapoint in that file
-agent-sdk eval --dir . weather forecast    # several files
+agent-sdk eval --dir . --list
+agent-sdk eval --dir . --json
+agent-sdk eval --dir . weather/nyc
 agent-sdk eval --dir . --tag smoke
-agent-sdk eval --dir . --verbose
 ```
 
-`eval` (like `run`) boots an ephemeral server on port 0 with a temp state
-root outside the project, so cases don't inherit ambient monorepo rules and
-don't pollute `.agent-serve/`. Point `--url` at a running server instead to
-eval a live agent. Model turns require `CURSOR_API_KEY`.
-
-## Seeding / adding samples (required fork)
+| Form | Case id |
+| --- | --- |
+| `evals/weather.eval.ts` + `test` | `weather` |
+| `evals/weather/nyc.eval.ts` + `test` | `weather/nyc` |
+| `evals/weather.eval.ts` + `{ id: "nyc" }` | `weather/nyc` |
 
-Whenever the task is to **create, seed, or expand** eval cases (not merely
-run or tweak an existing gate), start with the `AskQuestion` tool. Do not
-assume a source of fixtures.
+`eval` boots an ephemeral server and a temp state root. `--url`
+points at a running agent. Model turns need `CURSOR_API_KEY`.
 
-### Step 1 — who supplies the samples?
+## Seeding
 
-Use `AskQuestion` (single-select). Put the recommended option first and
-label it `(recommended)`. Do **not** put code blocks inside the question
-prompt (they render poorly); keep the prompt short plain text.
+Creating or expanding cases: `AskQuestion` first.
 
 | Question | Options |
 | --- | --- |
 | **How should we get eval samples?** | Generate test eval samples for me `(recommended)` · I will add / upload the data manually |
 
-If `AskQuestion` is unavailable, ask the same fork in plain chat and wait.
-
-### Step 2a — user chose manual
-
-Tell the user the **shape** of the data (see “The API” below) and **how to
-upload it**. Do not invent cases until they provide files (or paste content
-to write). After they add data, run `agent-sdk eval --dir <project>
---list` to confirm discovery, then wire any missing gates.
-
-If the uploaded data is **API-backed** (PR URLs, pinned SHAs, gold labels,
-snapshot ids) rather than self-contained prompts/diffs, **materialize
-fixtures first** — see “When eval data needs API calls to render” below.
-
-**How to upload:**
-
-1. Add or edit `.eval.ts` files under the project’s `evals/` directory.
-2. Ensure `evals/evals.config.ts` exists with `maxConcurrency`.
-3. Optional fixtures beside the case or under `fixtures/`.
-4. Tell the agent once the files are in place — or paste case bodies in chat
-   and ask the agent to write the file.
-5. Verify: `agent-sdk eval --dir <project> --list` shows the new ids.
-
-### Step 2b — user chose agent-generated samples
-
-1. Use `AskQuestion` again for count (single-select), recommended first:
-
-   | Question | Options |
-   | --- | --- |
-   | **How many test samples should I create?** | `3` `(recommended)` · `5` · `10` · Other (enter a number) |
-
-2. Inspect the agent (tools, instructions, existing `evals/`) and invent
-   that many **distinct** datapoints: frozen prompts + decision/shape gates,
-   not exact prose. Follow “What good cases assert” below.
-3. **Append** — prefer adding entries to an existing file’s `cases` array
-   when the suite fits; otherwise create a new `evals/<suite>.eval.ts` with
-   `cases`. Never overwrite or weaken an existing datapoint. Case ids must
-   not collide with `agent-sdk eval --dir <project> --list`.
-4. Create `evals/evals.config.ts` if missing (`maxConcurrency: 20` is fine
-   for now; hard limit 200 for model provider request limits).
-5. Re-list to confirm, then optionally run `--tag smoke` if you tagged any.
+1. **Manual.** They provide files or paste. Show the shape below.
+   Do not invent cases. Then `eval --list` and wire gates.
+   API-backed pointers (PR URLs, SHAs, gold labels): materialize
+   under `fixtures/` first.
+2. **Generated.** Ask count (`3` recommended). Append to an
+   existing `cases` array when it fits. Never overwrite or weaken
+   a datapoint. Create `evals/evals.config.ts` if missing
+   (`maxConcurrency: 20`; cap 200).
 
-## The API
+## API
 
 ```ts
-import { defineEval, includes, equals, satisfies } from "@cursor/july/evals";
+import { defineEval, includes, satisfies } from "@cursor/july/evals";
 
-// Single datapoint (case id = file path under evals/)
-export default defineEval({
-  description: "Uses get_weather for temperature questions.",
-  tags: ["smoke"],
-  timeoutMs: 120_000,
-  async test(t) {
-    await t.send("What's the temperature in NYC?");
-    t.succeeded();
-    t.calledTool("get_weather");
-    t.notCalledTool("save_weather_note");
-    t.messageIncludes(/°|[FC]/);
-    t.check(t.reply, includes("NYC"));
-  },
-});
-
-// Multiple datapoints in one file (ids = <fileId>/<case.id>)
 export default defineEval({
   tags: ["smoke", "weather"],
   cases: [
@@ -134,100 +63,46 @@ export default defineEval({
         await t.send("What's the temperature in NYC?");
         t.succeeded();
         t.calledTool("get_weather");
+        t.notCalledTool("save_weather_note");
         t.check(t.reply, includes(/°|[FC]/));
       },
     },
-    {
-      id: "paris",
-      async test(t) {
-        await t.send("What's the weather in Paris?");
-        t.succeeded();
-        t.calledTool("get_weather");
-      },
-    },
   ],
 });
 ```
 
-Project-wide defaults in `evals/evals.config.ts` (required once):
-
 ```ts
-import {
-  defineEvalConfig,
-  persistEvalRunsToDir,
-} from "@cursor/july/evals";
+import { defineEvalConfig } from "@cursor/july/evals";
 
 export default defineEvalConfig({
-  maxConcurrency: 20, // required; hard-capped at 200 (model provider limits)
-  // Playground batches are memory-only unless you set persistRuns:
-  // maxPlaygroundRuns: 50, // optional; default 20; playground history only
-  // persistRuns: persistEvalRunsToDir(".agent-serve/eval-runs"),
+  maxConcurrency: 20,
 });
 ```
 
-- Either a single `test(t)` **or** a `cases: [{ id, test }, …]` array — not
-  both. Prefer `cases` for multiple related datapoints.
-- `t.send(message, options?)` runs one turn and waits for park/fail.
-  Options: `workspaceDir`, `workspaceFiles` (`{ path: contents }` seeded
-  before the first turn). Multiple `t.send` calls share the session.
-- `t.reply` / `t.events` / `t.sessionId` for assertions and correlation.
-- `t.metric(name, value)` — structured scores on the case snapshot (playground
-  badges + CLI). Prefer this over encoding scores only in `t.log`.
-- Builders: `includes(string | RegExp)`, `equals(value)`,
-  `satisfies(predicate, label)`.
-
-### Playground customization (optional)
-
-Drop `agent/playground/evals.tsx` exporting `agentName` (must match
-`defineAgent({ name })`) and
-`default: { RunSummary?, CaseMetrics?, CaseExtras? }`. The playground
-loads these via Vite glob (`factory/*`). Without a
-module, cases still show default `t.metric` badges.
-
-## What good cases assert
-
-Gate **decisions and shape, not prose**. Model wording varies run to run;
-tool choice, tool avoidance, and output structure shouldn't.
-
-1. `t.succeeded()` — always, first.
-2. The tool decision: `calledTool` for the intended path, `notCalledTool`
-   for the tempting wrong one (the pair is stronger than either alone).
-3. Output shape: a regex for the contract (`/°|[FC]/`, a JSON marker, a
-   findings-block fence) — not exact sentences.
-4. For structured output, parse `t.reply` and `t.check` fields with
-   `satisfies` rather than substring-matching JSON.
-5. When a shape gate keeps failing on model formatting, tighten the output
-   contract in instructions and re-run — do not move rendering into a host
-   tool to make the gate pass. That trades a prompt fix for permanent code.
-
-Anti-patterns: asserting exact phrasing; more than ~5 gates in one case
-(split it); cases that depend on live external state that drifts (pin the
-input instead — see fixtures).
-
-## Fixture strategy by agent type
-
-| Agent surface | Fixture source |
-| --- | --- |
-| Chat / domain assistant | A canonical prompt string, chosen once and frozen |
-| Tool-heavy | `agent-sdk call <tool> --dir . --input '{...}'` first, then the prompt that should trigger it |
-| GitHub webhook | `github replay … --dry-run --out fixtures/github` (see `packages/agent-serve/skills/github/SKILL.md`) |
-| PR reviewer with host prep | A small fixed PR the team controls; assert on findings shape, not counts |
-| Workspace-dependent | `workspaceFiles` in `t.send` — never machine-local paths |
+Either `test(t)` or `cases`, not both. `t.send` waits for park/fail.
+`workspaceFiles` seeds the first turn. Assert with `t.succeeded()`,
+`calledTool` / `notCalledTool`, `t.check(t.reply, …)`, `t.metric`.
 
-### When eval data needs API calls to render (rate limits)
+## What to gate
 
-If samples are **pointers** (PR URL, SHAs, `gold_labels`) rather than
-self-contained evidence, do **not** wire live `t.send(prUrl)` matrices.
-Materialize once under `fixtures/` at pinned SHAs, point cases at those
-files, gate on decision/shape against gold, and keep a tiny `--tag smoke`
-subset for any full-pipeline runs.
+Decisions and shape, not prose.
 
-## The ratchet rule (hillclimbing)
+1. `t.succeeded()` first
+2. Intended tool + the tempting wrong one
+3. A shape regex or `satisfies` on parsed fields
+4. If formatting keeps failing, tighten instructions. Do not move
+   rendering into a host tool.
 
-Every hillclimb round that **keeps** a change must land an eval that would
-have failed before the change. Never weaken a gate to make a round pass.
+Anti-patterns: exact phrasing; more than ~5 gates (split); live
+drifting inputs (pin them).
 
-```ts
-const toolResults = t.events.filter((e) => e.type === "action.result");
-t.check(toolResults.length, satisfies((n) => (n as number) <= 4, "≤4 tool calls"));
-```
+| Surface | Fixture |
+| --- | --- |
+| Chat | One frozen prompt |
+| Tool-heavy | `agent-sdk call` first, then the prompt |
+| GitHub | `github replay … --dry-run --out fixtures/github` |
+| Host-prep PR review | A team-owned PR; gate findings shape, not counts |
+| Workspace | `workspaceFiles` in `t.send` |
+
+Every kept hillclimb change lands an eval that would have failed
+before it. Never weaken a gate to pass a round.

package/skills/framework-map/SKILL.md

@@ -1,165 +1,103 @@
 ---
 name: agentsdk-framework-map
 description: >-
-  Orientation for @cursor/july — folder structure, local vs cloud
-  runtimes, sessions, state layout, and invariants (Node not Bun, root
-  evals/, typecheck after tsx). Use when creating, editing, or running
-  Agent SDK projects with agent/agent.ts and agent/instructions.md.
+  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.
 ---
 
 # Agent SDK framework map
 
-`@cursor/july` serves agents defined as ordinary files: markdown
-for prose, TypeScript for typed behavior, under an `agent/` directory. The
-framework discovers the files, compiles a manifest, and serves the agent
-over HTTP/Slack/GitHub channels, with the Cursor SDK + harness as the
-execution engine. Ground truth is the package `README.md` (full reference)
-and `AGENTS.md` (coding-agent loop) — this skill is the map, not the spec.
-
-Public docs index: `node_modules/@cursor/july/dist/docs/llms.txt`. A running
-host serves the same index at `/docs/llms.txt`.
-
-Run the CLI as `agent-sdk <cmd>`. In the everysphere monorepo there is no
-installed bin — use:
-
-```bash
-cd packages/agent-serve
-pnpm exec tsx src/bin/agent-serve.ts <cmd> ...
-```
-
-## Hard invariants (violating these wastes hours)
-
-1. **Node 22+, never Bun.** Bun's HTTP/2 client corrupts harness
-   tool-result streams (`NGHTTP2_FRAME_SIZE_ERROR`): every built-in
-   read/grep the model makes fails and turns degrade into 8-minute retry
-   loops. All package mise tasks already use tsx.
-2. **Evals live at the project root `evals/`**, never `agent/evals/` (that
-   path is silently ignored).
-3. **tsx strips types without checking them.** `validate`/`run` passing
-   does not mean the code typechecks — run the repo's TypeScript check
-   (in everysphere: `dev check typecheck packages/agent-serve/... --timeout 120`)
-   before shipping. Classic miss: a tool `execute` return that fails
-   `ToolExecuteResult` — returns must be JSON-shaped, so use object
-   literals or `type` aliases, not `interface` types.
-4. **Harness cwd in a nested git repo.** Discovery defaults `local.cwd` to
-   `~/.cache/agent-serve/<dir>` so ancestor `AGENTS.md` / `.cursor` / IDE
-   MCP do not leak into playground turns. Set cwd at a checkout only when
-   the agent must inherit that tree. Attached MCP is invisible by name
-   until `advertiseTools: true`.
-5. **A model turn needs `CURSOR_API_KEY`.** Everything structural
-   (`validate`, `info`, `call`, `serve` bring-up) works without one.
+`@cursor/july` discovers files under `agent/` and serves the agent
+over HTTP, Slack, and GitHub. Markdown is prose. TypeScript is typed
+behavior. Ground truth: package `README.md` and `AGENTS.md`.
+
+CLI is `agent-sdk` (Node, never Bun).
+
+Public docs: `node_modules/@cursor/july/dist/docs/llms.txt` or
+`/docs/llms.txt` on a running host.
+
+## Invariants
+
+1. **Node 22.13+, never Bun.** Bun corrupts harness tool-result
+   streams (`NGHTTP2_FRAME_SIZE_ERROR`).
+2. **Evals live at project-root `evals/`.** `agent/evals/` is ignored.
+3. **tsx does not typecheck.** Tool `execute` must return JSON-shaped
+   values: object literals or `type` aliases, not `interface` types.
+4. **Nested git checkouts.** Discovery sets `local.cwd` to a
+   per-project cache directory under `~/.cache`. Point cwd at a
+   checkout only when the agent must inherit that tree.
+5. **Attached MCP is nameless** until `advertiseTools: true`.
+6. **Model turns need `CURSOR_API_KEY`.** `validate`, `info`, `call`,
+   and `serve` bring-up do not.
 
 ## Folder structure
 
-One file per concept; the path is the identity. Full details for each
-path live under README "Folder structure".
+Path is identity. Full list: README "Folder structure".
 
-| Path | What it is |
+| Path | Role |
 | --- | --- |
-| `agent/agent.ts` | `defineAgent({ model?, runtime?, cloud?, local? })`; model defaults to `grok-4.5` with `effort=high`, `fast=true` |
-| `agent/instructions.md` | Always-on system prompt (required; `.ts`/dir forms exist) |
-| `agent/tools/<name>.ts` | One typed tool; filename = tool name. `execution: "server"` (in-process, default) or `"agent"` (script that runs where the agent runs) |
-| `agent/skills/*` | SKILL.md-convention procedures, loaded on demand |
-| `agent/mcp-connections/<name>.ts` | MCP servers. Do not name an account connection `cursor.ts`. `advertiseTools: true` for named tools on local turns. `{ tools }` authors TypeScript the Agent SDK packages as stdio MCP. |
-| `agent/host-connections/<name>.ts` | Privileged MCP for `ctx.host.mcp` and `mcp oauth`. The model never sees them. |
-| `agent/subagents/<id>/` | Child agent dir (`description` required; per-subagent tools/skills/MCP connections ignored for now) |
-| `agent/channels/*.ts` | HTTP surfaces beyond the built-in session API; `slack.ts` / `github.ts` use the platform packs |
-| `agent/hooks/*.ts` | Observe-only event subscribers (never fatal) |
-| `agent/ab.ts` or `agent/ab/*.ts` | Live A/B metrics (`defineAB`); sticky `split` + `onSample` |
-| `agent/ab.config.ts` | Optional A/B `maxPlaygroundSessions` / `persistSamples` / `persistSnapshots` |
-| `agent/otel.ts` | OpenTelemetry export (`defineOtel`); also `OTEL_EXPORTER_OTLP_*` env |
-| `agent/schedules/*` | Cron-driven runs (UTC, 5-field; never auto-fire under `--dev`) |
-| `agent/sandbox/workspace/` | Files seeded into each session workspace (local runtime only) |
-| `agent/lib/` | Import-only shared code, never discovered |
-| `evals/**/*.eval.ts` | Filesystem evals; case id = path under `evals/` |
-
-## Local vs cloud runtime
-
-`runtime: "local"` (default) runs turns on the Cursor SDK local harness on
-the serve host; `runtime: "cloud"` runs them on Cursor cloud agents
-(`bc-...` ids, needs `cloud: { repos: [...] }`).
+| `agent/agent.ts` | `defineAgent({ model?, runtime?, cloud?, local? })` |
+| `agent/instructions.md` | Always-on system prompt (required) |
+| `agent/tools/<name>.ts` | One tool. `execution: "server"` or `"agent"` |
+| `agent/skills/*` | On-demand procedures |
+| `agent/mcp-connections/<name>.ts` | MCP. Never name an account file `cursor.ts`. `advertiseTools: true` for named local tools |
+| `agent/host-connections/<name>.ts` | Privileged MCP for `ctx.host.mcp` / `mcp oauth` |
+| `agent/subagents/<id>/` | Child agent (`description` required) |
+| `agent/channels/*.ts` | Slack / GitHub / custom HTTP |
+| `agent/hooks/*.ts` | Observe-only |
+| `agent/ab.ts` or `agent/ab/*.ts` | Live A/B (`defineAB`) |
+| `agent/otel.ts` | OpenTelemetry (`defineOtel`) |
+| `agent/schedules/*` | Cron. Never auto-fire under `--dev` |
+| `agent/sandbox/workspace/` | Session seed files (local only) |
+| `agent/lib/` | Import-only. Never discovered |
+| `evals/**/*.eval.ts` | Case id is the path under `evals/` |
+
+## Local vs cloud
+
+`runtime: "local"` (default) runs on the serve host.
+`runtime: "cloud"` needs `cloud: { repos: [...] }`.
 
 | Capability | local | cloud |
 | --- | --- | --- |
-| Server tools (`execution: "server"`) | yes | yes on managed hosting; self-hosted needs `--public-url` |
+| Server tools | yes | yes on managed hosting; self-hosted needs `--public-url` |
 | Tool approvals | yes | no |
-| Agent tools (`execution: "agent"` scripts) | yes | yes |
-| Skills | yes | yes |
+| Agent tools / skills | yes | yes |
 | sandbox seeds | yes | no |
-| Instructions | `AGENTS.md` in session workspace | prepended to first prompt |
-| Checkout of a repo/PR | you arrange it | the VM carries it |
-
-Rule of thumb: per-PR worktrees on the serve host die at
-"hundreds of engineers issuing PRs" — when the job needs a repo checkout at
-scale, use cloud and keep the host to routing, briefs, and bookkeeping.
-`validate` warns when cloud is combined with local-only capabilities.
-
-## Sessions, tokens, streams
-
-Two handles, two owners:
-
-- **continuationToken:** credential for continuing a conversation. HTTP
-  follow-ups rotate it; stale tokens get `409`.
-- **sessionId** — runtime-owned handle for streaming/inspection
-  (`GET /v1/session/:id/stream?startIndex=N`, durable NDJSON replay).
-
-A follow-up to a busy HTTP/MCP session **interrupts the in-flight turn**
-and waits for it to settle before the new turn sends. Slack uses
-**coalesce** instead (enqueue + boundary drain; see
-`docs/reference/sessions.md`). `POST /v1/session/:id/stop` interrupts
-without sending. Session-bound deterministic tool calls are rejected with
-`409 session_busy` while a turn runs.
-
-Key built-in routes (each agent, under `/<slug>` in multi-agent mode):
-`POST /v1/session`, `POST /v1/session/:id`, `GET /v1/session/:id/stream`,
-`GET /v1/sessions`, `POST /v1/tools/:toolName` (deterministic call),
-`GET/POST /v1/session/:id/approvals[/:callId]`, `GET /v1/info`,
-`GET /v1/health`, and in dev `POST /v1/dev/schedules/:id`,
-`POST /v1/dev/reminders/:id`.
-
-## State layout
-
-```
-<project>/.agent-serve/            # or <stateRoot>/<slug>/ under serve
-  sessions/<id>/session.json       # metadata
-  sessions/<id>/events.ndjson      # durable stream (replayable)
-  sessions/<id>/workspace/         # the harness cwd for that session
-  traces/<sessionId>.ndjson        # written by `run`
-  runner/                          # SDK conversation store
-```
-
-Delete a session directory to forget a conversation.
+| Checkout | you arrange it | the VM carries it |
+
+Use cloud when the job needs a checkout at scale. `validate` warns
+when cloud is combined with local-only capabilities.
+
+## Sessions
+
+- **continuationToken** continues a conversation. HTTP follow-ups
+  rotate it. Stale tokens return `409`.
+- **sessionId** is the inspect handle
+  (`GET /v1/session/:id/stream?startIndex=N`).
+
+A follow-up to a busy HTTP/MCP session interrupts the in-flight
+turn. Slack coalesces. Routes: `docs/reference/http-api.md`.
+Session files live under the project state directory
+(`--state-root`).
 
 ## Where logic belongs
 
-The model is the engine; TypeScript is the boundary layer. Code earns its
-place on the deterministic path for side-effect gates and actuation,
-idempotency/dedupe of external writes, auth/secrets, evidence seeding, and
-hard caps at API boundaries. Formatting, summarizing, classification,
-wording, and composing human-facing output belong to the model, under an
-output contract in instructions/skills — prose lives in `.md` files, not TS
-string builders. When `agent/lib/` + `tools/` dwarf instructions + skills,
-you are usually looking at a service with a vestigial model turn, or at
-framework gaps patched agent-side that should be fixed upstream. The
-budget table and smells: `create-agent/SKILL.md` § "The deterministic-path
-budget".
-
-## The loop
-
-Edit → inspect → run → assert. Canonical commands and flags live in
-`AGENTS.md` (`validate` / `info` / `call` / `run` / `eval` / `serve`).
-Serve only your agent's directory during bring-up — mounting a parent
-folder mounts every sibling agent too.
-
-## Where to go next
-
-- Scaffolding → `packages/agent-serve/skills/create-agent/SKILL.md`
-- Evals → `packages/agent-serve/skills/evals/SKILL.md`
-- Live A/B metrics → `packages/agent-serve/skills/ab/SKILL.md`
-- OpenTelemetry → `packages/agent-serve/skills/otel/SKILL.md`
-- GitHub webhooks → `packages/agent-serve/skills/github/SKILL.md`
-- Slack → `packages/agent-serve/skills/setup-slack/SKILL.md`
-- Host MCP OAuth → `packages/agent-serve/skills/mcp-auth/SKILL.md`
-- Local triage → `packages/agent-serve/skills/debug/SKILL.md`
-- Measured improvement → `packages/agent-serve/skills/hillclimb/SKILL.md`
-- Dogfood / deploy → `packages/agent-serve/skills/operate/SKILL.md`
+Code: side-effect gates, write dedupe, auth, evidence seeding, hard
+API caps. Model: formatting, summarizing, classification, replies.
+Budget: `skills/create-agent/SKILL.md`.
+
+Loop: `validate` / `info` / `call` / `run` / `eval` / `serve`. Serve
+only this agent's directory.
+
+| Task | Skill |
+| --- | --- |
+| Scaffold | `skills/create-agent/SKILL.md` |
+| Evals | `skills/evals/SKILL.md` |
+| Live A/B | `skills/ab/SKILL.md` |
+| OpenTelemetry | `skills/otel/SKILL.md` |
+| GitHub | `skills/github/SKILL.md` |
+| Slack | `skills/setup-slack/SKILL.md` |
+| Host MCP OAuth | `skills/mcp-auth/SKILL.md` |
+| Local triage | `skills/debug/SKILL.md` |
+| Measured improvement | `skills/hillclimb/SKILL.md` |