Import pelikhan/rig into githubnext/rig and harden JSON fallback parsing

.gitattributes

@@ -0,0 +1 @@
+.github/workflows/*.lock.yml linguist-generated=true merge=ours

.github/mcp.json

@@ -0,0 +1,11 @@
+{
+  "mcpServers": {
+    "github-agentic-workflows": {
+      "command": "gh",
+      "args": [
+        "aw",
+        "mcp-server"
+      ]
+    }
+  }
+}

.github/workflows/ci.yml

@@ -0,0 +1,51 @@
+name: CI
+on:
+  push:
+  pull_request:
+permissions:
+  contents: read
+jobs:
+  build-test:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Use Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: 24
+          cache: npm
+
+      - name: Install dependencies
+        run: npm ci
+
+      - name: Typecheck
+        run: npm run typecheck
+
+      - name: Test
+        run: npm test
+
+  integration-test:
+    runs-on: ubuntu-latest
+    needs: build-test
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Use Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: 24
+          cache: npm
+
+      - name: Install dependencies
+        run: npm ci
+
+      - name: Integration test
+        env:
+          COPILOT_GITHUB_TOKEN: ${{ secrets.COPILOT_GITHUB_TOKEN }}
+        run: |
+          if [ -z "${COPILOT_GITHUB_TOKEN:-}" ]; then
+            echo "Skipping integration tests: COPILOT_GITHUB_TOKEN is not available."
+          else
+            npm run test:integration
+          fi

.github/workflows/copilot-setup-steps.yml

@@ -0,0 +1,26 @@
+name: "Copilot Setup Steps"
+
+# This workflow configures the environment for GitHub Copilot Agent with gh-aw MCP server
+on:
+  workflow_dispatch:
+  push:
+    paths:
+      - .github/workflows/copilot-setup-steps.yml
+
+jobs:
+  # The job MUST be called 'copilot-setup-steps' to be recognized by GitHub Copilot Agent
+  copilot-setup-steps:
+    runs-on: ubuntu-latest
+
+    # Set minimal permissions for setup steps
+    # Copilot Agent receives its own token with appropriate permissions
+    permissions:
+      contents: read
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v6
+      - name: Install gh-aw extension
+        uses: github/gh-aw-actions/setup-cli@v0.77.5
+        with:
+          version: v0.77.5

.github/workflows/daily-rig-sampler.md

@@ -0,0 +1,97 @@
+---
+name: Daily Rig Sampler
+description: >
+  Each day, pick one rig sample (cached round-robin) from src/samples/,
+  run it with the rig skill, analyze harness performance, and apply one
+  focused quick-win improvement to skills/rig/rig.ts in a new draft PR.
+on:
+  schedule: daily
+  workflow_dispatch:
+permissions:
+  contents: read
+  actions: read
+  issues: read
+  pull-requests: read
+engine: copilot
+strict: true
+timeout-minutes: 30
+tools:
+  github:
+    mode: gh-proxy
+    toolsets: [default]
+  bash: ["*"]
+  edit:
+  cache-memory: true
+network:
+  allowed: [defaults, github, node]
+safe-outputs:
+  create-pull-request:
+    title-prefix: "[rig-sampler] "
+    labels: [automation, ai-agent]
+    draft: true
+    reviewers: [copilot]
+    allowed-files:
+      - "skills/rig/rig.ts"
+---
+
+## Task
+
+You are improving the rig TypeScript harness one focused change at a time.
+
+### Step 1 — Pick the next sample (cached round-robin)
+
+1. Run `ls src/samples/` to list all sample files sorted alphabetically.
+2. Open the cache-memory file `/tmp/gh-aw/cache-memory/last-sample.json`.
+   - If it does not exist, start from the first file.
+   - Otherwise read the `lastFile` field and advance to the next file in the sorted list (wrapping around).
+3. Write `{"lastFile": "<chosen-file>"}` back to `/tmp/gh-aw/cache-memory/last-sample.json`.
+4. Note the chosen sample file path, e.g. `src/samples/05-write-readme-intent.ts`.
+
+### Step 2 — Install dependencies and run the sample
+
+1. Run `npm install` in the repository root to ensure all dependencies are present.
+2. Run the chosen sample through the rig skill launcher:
+   ```
+   node skills/rig/rig.ts <chosen-sample-path> 2>&1
+   ```
+   Capture both stdout (the agent's structured JSON output) and stderr (JSONL event lines prefixed with `rig.copilot-ask`).
+3. Record the full output for analysis.
+
+### Step 3 — Analyze harness performance
+
+With the captured run output, evaluate:
+
+- **Did it succeed?** Did the agent return valid JSON matching the declared output schema, or did it fail?
+- **Repair turns**: Count how many times the harness had to retry due to invalid JSON or schema violations (`rig.copilot-ask` events with `turns > 1`).
+- **Turn count and latency**: Note total turns from the JSONL events.
+- **Schema fit**: Did the sample's output schema feel too loose (e.g. plain `s.string` where a structured type would help) or too strict (repair loops due to enum mismatches)?
+- **Error messages**: Were any error messages unclear or unhelpful?
+- **API ergonomics**: Was there anything awkward in how the sample had to express its intent — boilerplate that a helper could eliminate, or a missing convenience on `p.*` or `s.*`?
+
+### Step 4 — Read rig.ts
+
+Read `skills/rig/rig.ts` in full to understand the current implementation.
+
+### Step 5 — Identify one quick-win improvement
+
+Based on your analysis of the actual run, identify **exactly one** small, self-contained improvement to `skills/rig/rig.ts`.
+
+Good categories (pick the one most directly supported by the run evidence):
+- A missing `s.*` schema helper that would simplify sample code or prevent a repair loop.
+- A clearer error message surfaced during repair or schema validation.
+- A JSDoc comment on a public export that is currently undocumented.
+- A small type-safety improvement (stricter overload, narrower generic).
+- A minor performance or usability tweak with no behaviour change.
+
+Do **not** change the public API in a breaking way. Keep the change small and reviewable.
+
+### Step 6 — Apply the improvement
+
+Edit `skills/rig/rig.ts` to implement the improvement. Use the `edit` tool.
+
+### Step 7 — Create a pull request
+
+Emit a `create-pull-request` output with:
+- `title`: one-line description of the improvement (no prefix needed — it is added automatically).
+- `body`: explain which sample was run, what the run revealed (repair turns, output quality, etc.), and why this change improves the harness.
+- `branch`: `rig-sampler/<chosen-sample-basename>` (e.g. `rig-sampler/05-write-readme-intent`).

.github/workflows/release.yml

@@ -0,0 +1,107 @@
+name: Release
+
+on:
+  workflow_dispatch:
+    inputs:
+      release_type:
+        description: "Release type"
+        required: true
+        type: choice
+        options:
+          - patch
+          - minor
+          - major
+
+jobs:
+  release:
+    runs-on: ubuntu-latest
+    # Only run on the origin repo (not forks)
+    if: github.repository == 'pelikhan/rig'
+    permissions:
+      contents: write
+
+    steps:
+      - name: Check actor permission
+        uses: actions/github-script@3a2844b7e9c422d3c10d287c895573f7108da1b3 # v9.0.0
+        with:
+          script: |
+            const { data: perm } = await github.rest.repos.getCollaboratorPermissionLevel({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              username: context.actor,
+            });
+            const level = perm.permission;
+            if (level !== 'admin' && level !== 'maintain') {
+              core.setFailed(`Actor ${context.actor} has permission '${level}', which is insufficient. Admin or maintain required.`);
+            }
+
+      - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
+        with:
+          fetch-depth: 0
+
+      - name: Use Node.js
+        uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6.4.0
+        with:
+          node-version: 24
+          cache: npm
+
+      - name: Install dependencies
+        run: npm ci
+
+      - name: Typecheck
+        run: npm run typecheck
+
+      - name: Compute next version
+        id: semver
+        run: |
+          # Get the latest semver tag (vX.Y.Z), default to v0.0.0 if none exists
+          LAST_TAG=$(git tag --list 'v[0-9]*.[0-9]*.[0-9]*' --sort=-version:refname | head -n 1)
+          if [ -z "$LAST_TAG" ]; then
+            LAST_TAG="v0.0.0"
+          fi
+          echo "Last tag: $LAST_TAG"
+
+          # Strip leading 'v'
+          VERSION="${LAST_TAG#v}"
+          MAJOR=$(echo "$VERSION" | cut -d. -f1)
+          MINOR=$(echo "$VERSION" | cut -d. -f2)
+          PATCH=$(echo "$VERSION" | cut -d. -f3)
+
+          RELEASE_TYPE="${{ github.event.inputs.release_type }}"
+          case "$RELEASE_TYPE" in
+            major)
+              MAJOR=$((MAJOR + 1))
+              MINOR=0
+              PATCH=0
+              ;;
+            minor)
+              MINOR=$((MINOR + 1))
+              PATCH=0
+              ;;
+            patch)
+              PATCH=$((PATCH + 1))
+              ;;
+          esac
+
+          NEW_TAG="v${MAJOR}.${MINOR}.${PATCH}"
+          echo "New tag: $NEW_TAG"
+          echo "tag=$NEW_TAG" >> "$GITHUB_OUTPUT"
+
+      - name: Create and push tag
+        run: |
+          git config user.name "github-actions[bot]"
+          git config user.email "github-actions[bot]@users.noreply.github.com"
+          git tag "${{ steps.semver.outputs.tag }}"
+          git push origin "${{ steps.semver.outputs.tag }}"
+
+      - name: Create GitHub Release
+        uses: actions/github-script@3a2844b7e9c422d3c10d287c895573f7108da1b3 # v9.0.0
+        with:
+          script: |
+            await github.rest.repos.createRelease({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              tag_name: '${{ steps.semver.outputs.tag }}',
+              name: '${{ steps.semver.outputs.tag }}',
+              generate_release_notes: true,
+            });

.gitignore

@@ -0,0 +1 @@
+node_modules/

.vscode/settings.json

@@ -0,0 +1,5 @@
+{
+    "github.copilot.enable": {
+        "markdown": true
+    }
+}

AGENTS.md

@@ -0,0 +1,61 @@
+# AGENTS.md
+
+## Project Overview
+
+Rig is a minimal TypeScript agent harness. The core runtime (`skills/rig/rig.ts`) provides declarative agent construction with typed input/output shapes, prompt intents, and a Copilot SDK runtime.
+
+## Architecture
+
+```
+skills/rig/rig.ts      — Core runtime (agent, p, copilotEngine, schemas)
+skills/rig/samples/    — 51 sample agents demonstrating patterns
+src/engines/copilot.test.ts — Copilot engine unit tests (vitest)
+src/rig.test.ts        — Unit tests (vitest)
+scripts/run-sample.test.ts — Sample runner with a stub Copilot SDK client (dry-run)
+skills/rig/SKILL.md    — Framework reference docs
+```
+
+All imports use the `"rig"` path alias (resolved via tsconfig paths + vitest alias). `copilotEngine` is exported directly from `rig` for client construction.
+
+## Commands
+
+| Task | Command |
+|------|---------|
+| Typecheck | `npm run typecheck` |
+| Unit tests | `npm test` |
+| Run samples (stub) | `npm run sample` |
+| Run single sample (stub) | `RIG_SAMPLE=02 npm run sample` |
+| Run a sample for real | `echo "<input>" \| node skills/rig/rig.ts <program-file>` (`npm run sample:run`) |
+
+## Code Style
+
+- Keep the core (`skills/rig/rig.ts`) self-contained; `@github/copilot-sdk` is imported directly in `skills/rig/rig.ts`
+- Minimal comments; code should be self-explanatory
+- Use `node:` prefix for Node.js built-in imports
+- Types are colocated with the module that defines them, not in separate `.d.ts` files
+- Trailing underscore on object keys (`key_`) means optional field
+- Do not add legacy compatibility bridges; update callers, samples, and docs to the current API
+
+## Testing
+
+- Framework: vitest
+- Tests live in `src/rig.test.ts` (agent definition, invocation, validation, and prompt intent coverage)
+- Stub the Copilot SDK client with `vi.mock("@github/copilot-sdk", ...)`
+- All unit tests must pass before committing
+- Samples run via a stub Copilot SDK client that synthesizes shape-conforming output from the prompt's `<output_schema>` block
+
+## Key Concepts
+
+- **Shape descriptors**: JS values used as type exemplars (e.g., `""` = string, `0` = number, `[""]` = string array). Promoted to schemas via `SchemaLike`.
+- **Schema helpers (`s.*`)**: `s.string`, `s.number`, `s.boolean`, `s.unknown`, `s.array`, `s.object`, `s.record`, `s.enum`, `s.optional`
+- **Prompt intents (`p.*`)**: `p.bash(cmd)`, `p.read(path)`, `p.write(path, content)` — declarative placeholders resolved into prompt instructions, not executed in-process
+- **Prompts**: `p\`...\`` template tag composes instructions with inline `p.*` helpers
+- **Runtime transport**: Copilot SDK sessions are created by the harness; use launcher `--server` to switch to stdio transport.
+- **Repair**: built-in addon re-prompts on parse/validation failure up to `maxTurns`, and other addons can still steer retry prompts.
+
+## Sample guide
+
+- `20-issue-reproducer.ts` — chained diagnosis, fix planning, and review
+- `36-subagent-delegation.ts` — focused-agent delegation
+- `47-prompt-intents.ts` — prompt intent primitives
+- `50-end-to-end-release-agent.ts` — end-to-end release workflow orchestration