docs: add Copilot + @claude agent configuration

.github/copilot-instructions.md

@@ -0,0 +1,45 @@
+# ShellForge — Copilot Instructions
+
+> Copilot acts as **Tier C — Execution Workforce** in this repository.
+> Implement well-specified issues, open draft PRs, never merge or approve.
+
+## Project Overview
+
+**ShellForge** is the execution harness for the Chitin platform — agent loop, LLM providers, tool-use, drift detection.
+
+## Tech Stack
+
+- **Language**: Go 1.22+
+- **Module**: `github.com/chitinhq/shellforge`
+
+## Build & Test
+
+```bash
+go build ./...
+go test ./...
+golangci-lint run
+```
+
+## Governance Rules
+
+### DENY
+- `git push` to main — always use feature branches
+- `git force-push` — never rewrite shared history
+- Write to `.env`, SSH keys, credentials
+
+### ALWAYS
+- Create feature branches: `agent/<type>/issue-<N>`
+- Run `go build ./... && go test ./...` before creating PRs
+- Link PRs to issues (`Closes #N`)
+
+## PR Rules
+
+- **NEVER merge PRs** — only Tier B or humans merge
+- Max 300 lines changed per PR (soft limit)
+- Always open as **draft PR** first
+- If ambiguous, label `needs-spec` and stop
+
+## Autonomy Directive
+
+- **NEVER pause to ask for clarification** — make your best judgment
+- If the issue is ambiguous, label it `needs-spec` and stop

.github/workflows/claude-code-review.yml

@@ -0,0 +1,29 @@
+name: Claude Code Review
+
+on:
+  pull_request:
+    types: [opened, synchronize, ready_for_review, reopened]
+
+jobs:
+  claude-review:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: read
+      issues: read
+      id-token: write
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
+
+      - name: Run Claude Code Review
+        id: claude-review
+        uses: anthropics/claude-code-action@v1
+        with:
+          claude_code_oauth_token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
+          plugin_marketplaces: 'https://github.com/anthropics/claude-code.git'
+          plugins: 'code-review@claude-code-plugins'
+          prompt: '/code-review:code-review ${{ github.repository }}/pull/${{ github.event.pull_request.number }}'

.github/workflows/claude.yml

@@ -0,0 +1,39 @@
+name: Claude Code
+
+on:
+  issue_comment:
+    types: [created]
+  pull_request_review_comment:
+    types: [created]
+  issues:
+    types: [opened, assigned]
+  pull_request_review:
+    types: [submitted]
+
+jobs:
+  claude:
+    if: |
+      (github.event_name == 'issue_comment' && contains(github.event.comment.body, '@claude')) ||
+      (github.event_name == 'pull_request_review_comment' && contains(github.event.comment.body, '@claude')) ||
+      (github.event_name == 'pull_request_review' && contains(github.event.review.body, '@claude')) ||
+      (github.event_name == 'issues' && (contains(github.event.issue.body, '@claude') || contains(github.event.issue.title, '@claude')))
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      pull-requests: read
+      issues: read
+      id-token: write
+      actions: read
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 1
+
+      - name: Run Claude Code
+        id: claude
+        uses: anthropics/claude-code-action@v1
+        with:
+          claude_code_oauth_token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
+          additional_permissions: |
+            actions: read

CLAUDE.md

@@ -0,0 +1,22 @@
+## Agent Identity
+
+At session start, if you see `[AgentGuard] No agent identity set`, ask the user:
+1. **Role**: developer / reviewer / ops / security / planner
+2. **Driver**: human / claude-code / copilot / ci
+
+Then run: `scripts/write-persona.sh <driver> <role>`
+
+## Project
+
+ShellForge is the execution harness for the Chitin platform — agent loop, LLM providers, tool-use, drift detection.
+
+**Module**: `github.com/chitinhq/shellforge`
+**Language**: Go 1.22+
+
+## Build
+
+```bash
+go build ./...
+go test ./...
+golangci-lint run
+```