feat: add GitHub Copilot CLI support

AGENTS.md

@@ -44,6 +44,6 @@ bun run skill:check      # health dashboard for all skills
 ## Key conventions
 
 - SKILL.md files are **generated** from `.tmpl` templates. Edit the template, not the output.
-- Run `bun run gen:skill-docs --host codex` to regenerate Codex-specific output.
+- Run `bun run gen:skill-docs --host codex` to regenerate Codex output (Copilot derives from this at setup time).
 - The browse binary provides headless browser access. Use `$B <command>` in skills.
 - Safety skills (careful, freeze, guard) use inline advisory prose — always confirm before destructive operations.

CHANGELOG.md

@@ -1,5 +1,21 @@
 # Changelog
 
+## [0.11.10.0] - 2026-03-23 — Copilot CLI Support
+
+### Added
+
+- **gstack now works with GitHub Copilot CLI.** Run `./setup --host copilot` and all 28 skills install to `~/.copilot/skills/`. Works the same as Codex — skills live in `.agents/skills/` and are discovered automatically.
+- **Auto-detection finds Copilot CLI.** `./setup --host auto` now detects `copilot` alongside Claude Code, Codex, and Kiro. Install once, works everywhere.
+- **Session discovery includes Copilot.** The global discover tool (`bin/gstack-global-discover.ts`) scans `~/.copilot/session-state/` so `/retro` and cross-project dashboards count Copilot sessions.
+- **Health checks cover Copilot.** `bun run skill:check` and CI now verify Copilot skill freshness alongside Claude and Codex.
+
+### For contributors
+
+- Added `'copilot'` host type to `gen-skill-docs.ts`, `setup`, `gstack-global-discover.ts`, and `skill-check.ts`.
+- New E2E test infrastructure: `copilot-e2e.test.ts` and `copilot-session-runner.ts` paralleling Codex equivalents.
+- Updated `CONTRIBUTING.md` "Dual-host" → "Multi-host" with Copilot generation commands and testing guidance.
+- Targets the standalone GA Copilot CLI (`copilot` binary via `npm install -g @github/copilot`), not the legacy `gh copilot` extension.
+
 ## [0.11.9.0] - 2026-03-23 — Codex Skill Loading Fix
 
 ### Fixed

CLAUDE.md

@@ -22,6 +22,7 @@ bun run eval:summary # aggregate stats across all eval runs
 
 `test:evals` requires `ANTHROPIC_API_KEY`. Codex E2E tests (`test/codex-e2e.test.ts`)
 use Codex's own auth from `~/.codex/` config — no `OPENAI_API_KEY` env var needed.
+Copilot E2E tests (`test/copilot-e2e.test.ts`) require the standalone Copilot CLI (`npm install -g @github/copilot`).
 E2E tests stream progress in real-time (tool-by-tool via `--output-format stream-json
 --verbose`). Results are persisted to `~/.gstack-dev/evals/` with auto-comparison
 against the previous run.

CONTRIBUTING.md

@@ -213,11 +213,11 @@ SKILL.md files are **generated** from `.tmpl` templates. Don't edit the `.md` di
 # 1. Edit the template
 vim SKILL.md.tmpl              # or browse/SKILL.md.tmpl
 
-# 2. Regenerate for both hosts
+# 2. Regenerate for all hosts
 bun run gen:skill-docs
 bun run gen:skill-docs --host codex
 
-# 3. Check health (reports both Claude and Codex)
+# 3. Check health (reports Claude and Codex; Copilot derives from Codex at setup time)
 bun run skill:check
 
 # Or use watch mode — auto-regenerates on save
@@ -228,17 +228,17 @@ For template authoring best practices (natural language over bash-isms, dynamic
 
 To add a browse command, add it to `browse/src/commands.ts`. To add a snapshot flag, add it to `SNAPSHOT_FLAGS` in `browse/src/snapshot.ts`. Then rebuild.
 
-## Dual-host development (Claude + Codex)
+## Multi-host development (Claude + Codex + Copilot)
 
-gstack generates SKILL.md files for two hosts: **Claude** (`.claude/skills/`) and **Codex** (`.agents/skills/`). Every template change needs to be generated for both.
+gstack generates SKILL.md files for multiple hosts: **Claude** (`.claude/skills/`), **Codex** (`.agents/skills/`), and **Copilot** (`.agents/skills/`). Codex and Copilot share the same `.agents/skills/` output directory — only `--host codex` needs to be run during generation. Copilot-specific paths are rewritten at `setup --host copilot` time via sed. Every template change needs to be generated for Claude and Codex.
 
-### Generating for both hosts
+### Generating for all hosts
 
 ```bash
 # Generate Claude output (default)
 bun run gen:skill-docs
 
-# Generate Codex output
+# Generate Codex output (Copilot shares this — paths are rewritten at setup time)
 bun run gen:skill-docs --host codex
 # --host agents is an alias for --host codex
 
@@ -248,37 +248,37 @@ bun run build
 
 ### What changes between hosts
 
-| Aspect | Claude | Codex |
-|--------|--------|-------|
-| Output directory | `{skill}/SKILL.md` | `.agents/skills/gstack-{skill}/SKILL.md` (generated at setup, gitignored) |
-| Frontmatter | Full (name, description, allowed-tools, hooks, version) | Minimal (name + description only) |
-| Paths | `~/.claude/skills/gstack` | `$GSTACK_ROOT` (`.agents/skills/gstack` in a repo, otherwise `~/.codex/skills/gstack`) |
-| Hook skills | `hooks:` frontmatter (enforced by Claude) | Inline safety advisory prose (advisory only) |
-| `/codex` skill | Included (Claude wraps codex exec) | Excluded (self-referential) |
+| Aspect | Claude | Codex | Copilot |
+|--------|--------|-------|---------|
+| Output directory | `{skill}/SKILL.md` | `.agents/skills/gstack-{skill}/SKILL.md` (generated at setup, gitignored) | `.agents/skills/gstack-{skill}/SKILL.md` (generated at setup, gitignored) |
+| Frontmatter | Full (name, description, allowed-tools, hooks, version) | Minimal (name + description only) | Minimal (name + description only) |
+| Paths | `~/.claude/skills/gstack` | `$GSTACK_ROOT` (`.agents/skills/gstack` in a repo, otherwise `~/.codex/skills/gstack`) | `$GSTACK_ROOT` (`.agents/skills/gstack` in a repo, otherwise `~/.copilot/skills/gstack`) |
+| Hook skills | `hooks:` frontmatter (enforced by Claude) | Inline safety advisory prose (advisory only) | Inline safety advisory prose (advisory only) |
+| `/codex` skill | Included (Claude wraps codex exec) | Excluded (self-referential) | Excluded (shares Codex output) |
 
-### Testing Codex output
+### Testing Codex and Copilot output
 
 ```bash
-# Run all static tests (includes Codex validation)
+# Run all static tests (includes Codex and Copilot validation)
 bun test
 
-# Check freshness for both hosts
+# Check freshness for all hosts
 bun run gen:skill-docs --dry-run
 bun run gen:skill-docs --host codex --dry-run
 
-# Health dashboard covers both hosts
+# Health dashboard covers all hosts
 bun run skill:check
 ```
 
 ### Dev setup for .agents/
 
-When you run `bin/dev-setup`, it creates symlinks in both `.claude/skills/` and `.agents/skills/` (if applicable), so Codex-compatible agents can discover your dev skills too. The `.agents/` directory is generated at setup time from `.tmpl` templates — it is gitignored and not committed.
+When you run `bin/dev-setup`, it creates symlinks in both `.claude/skills/` and `.agents/skills/` (if applicable), so Codex, Copilot, and other compatible agents can discover your dev skills too. The `.agents/` directory is generated at setup time from `.tmpl` templates — it is gitignored and not committed.
 
 ### Adding a new skill
 
-When you add a new skill template, both hosts get it automatically:
+When you add a new skill template, all hosts get it automatically:
 1. Create `{skill}/SKILL.md.tmpl`
-2. Run `bun run gen:skill-docs` (Claude output) and `bun run gen:skill-docs --host codex` (Codex output)
+2. Run `bun run gen:skill-docs` (Claude output) and `bun run gen:skill-docs --host codex` (Codex output — Copilot derives from this at setup time)
 3. The dynamic template discovery picks it up — no static list to update
 4. Commit `{skill}/SKILL.md` — `.agents/` is generated at setup time and gitignored
 

README.md

@@ -54,28 +54,29 @@ Open Claude Code and paste this. Claude does the rest.
 
 Real files get committed to your repo (not a submodule), so `git clone` just works. Everything lives inside `.claude/`. Nothing touches your PATH or runs in the background.
 
-### Codex, Gemini CLI, or Cursor
+### Codex, Gemini CLI, Copilot CLI, or Cursor
 
 gstack works on any agent that supports the [SKILL.md standard](https://github.com/anthropics/claude-code). Skills live in `.agents/skills/` and are discovered automatically.
 
 Install to one repo:
 
 ```bash
 git clone https://github.com/garrytan/gstack.git .agents/skills/gstack
-cd .agents/skills/gstack && ./setup --host codex
+cd .agents/skills/gstack && ./setup --host codex   # or --host copilot
 ```
 
-When setup runs from `.agents/skills/gstack`, it installs the generated Codex skills next to it in the same repo and does not write to `~/.codex/skills`.
+When setup runs from `.agents/skills/gstack`, it installs the generated Codex and Copilot skills next to it in the same repo and does not write to `~/.codex/skills` or `~/.copilot/skills`.
 
 Install once for your user account:
 
 ```bash
 git clone https://github.com/garrytan/gstack.git ~/gstack
-cd ~/gstack && ./setup --host codex
+cd ~/gstack && ./setup --host codex   # or --host copilot
 ```
 
 `setup --host codex` creates the runtime root at `~/.codex/skills/gstack` and
-links the generated Codex skills at the top level. This avoids duplicate skill
+links the generated Codex skills at the top level. `setup --host copilot` does
+the same at `~/.copilot/skills/gstack`. This avoids duplicate skill
 discovery from the source repo checkout.
 
 Or let setup auto-detect which agents you have installed:
@@ -85,7 +86,7 @@ git clone https://github.com/garrytan/gstack.git ~/gstack
 cd ~/gstack && ./setup --host auto
 ```
 
-For Codex-compatible hosts, setup now supports both repo-local installs from `.agents/skills/gstack` and user-global installs from `~/.codex/skills/gstack`. All 28 skills work across all supported agents. Hook-based safety skills (careful, freeze, guard) use inline safety advisory prose on non-Claude hosts.
+For Codex, Copilot, and other compatible hosts, setup supports both repo-local installs from `.agents/skills/gstack` and user-global installs from `~/.codex/skills/gstack` or `~/.copilot/skills/gstack`. All 28 skills work across all supported agents. Hook-based safety skills (careful, freeze, guard) use inline safety advisory prose on non-Claude hosts.
 
 ## See it work
 
@@ -156,7 +157,7 @@ Each skill feeds into the next. `/office-hours` writes a design doc that `/plan-
 | `/canary` | **SRE** | Post-deploy monitoring loop. Watches for console errors, performance regressions, and page failures. |
 | `/benchmark` | **Performance Engineer** | Baseline page load times, Core Web Vitals, and resource sizes. Compare before/after on every PR. |
 | `/document-release` | **Technical Writer** | Update all project docs to match what you just shipped. Catches stale READMEs automatically. |
-| `/retro` | **Eng Manager** | Team-aware weekly retro. Per-person breakdowns, shipping streaks, test health trends, growth opportunities. `/retro global` runs across all your projects and AI tools (Claude Code, Codex, Gemini). |
+| `/retro` | **Eng Manager** | Team-aware weekly retro. Per-person breakdowns, shipping streaks, test health trends, growth opportunities. `/retro global` runs across all your projects and AI tools (Claude Code, Codex, Gemini, Copilot). |
 | `/browse` | **QA Engineer** | Real Chromium browser, real clicks, real screenshots. ~100ms per command. |
 | `/setup-browser-cookies` | **Session Manager** | Import cookies from your real browser (Chrome, Arc, Brave, Edge) into the headless session. Test authenticated pages. |
 | `/autoplan` | **Review Pipeline** | One command, fully reviewed plan. Runs CEO → design → eng review automatically with encoded decision principles. Surfaces only taste decisions for your approval. |

VERSION

@@ -1 +1 @@
-0.11.9.0
+0.11.10.0

bin/gstack-global-discover.ts

@@ -1,6 +1,6 @@
 #!/usr/bin/env bun
 /**
- * gstack-global-discover — Discover AI coding sessions across Claude Code, Codex CLI, and Gemini CLI.
+ * gstack-global-discover — Discover AI coding sessions across Claude Code, Codex CLI, Gemini CLI, and Copilot CLI.
  * Resolves each session's working directory to a git repo, deduplicates by normalized remote URL,
  * and outputs structured JSON to stdout.
  *
@@ -17,15 +17,15 @@ import { homedir } from "os";
 // ── Types ──────────────────────────────────────────────────────────────────
 
 interface Session {
-  tool: "claude_code" | "codex" | "gemini";
+  tool: "claude_code" | "codex" | "gemini" | "copilot";
   cwd: string;
 }
 
 interface Repo {
   name: string;
   remote: string;
   paths: string[];
-  sessions: { claude_code: number; codex: number; gemini: number };
+  sessions: { claude_code: number; codex: number; gemini: number; copilot: number };
 }
 
 interface DiscoveryResult {
@@ -36,6 +36,7 @@ interface DiscoveryResult {
     claude_code: { total_sessions: number; repos: number };
     codex: { total_sessions: number; repos: number };
     gemini: { total_sessions: number; repos: number };
+    copilot: { total_sessions: number; repos: number };
   };
   total_sessions: number;
   total_repos: number;
@@ -440,7 +441,69 @@ function scanGemini(since: Date): Session[] {
   return sessions;
 }
 
-// ── Deduplication ──────────────────────────────────────────────────────────
+function scanCopilot(since: Date): Session[] {
+  // GitHub Copilot CLI (standalone) stores session data in ~/.copilot/session-state/
+  // Each session is a subdirectory: ~/.copilot/session-state/{session-id}/
+  // containing events.jsonl, workspace.yaml, and other session files.
+  const sessionStateDir = join(homedir(), ".copilot", "session-state");
+  if (!existsSync(sessionStateDir)) return [];
+
+  const sessions: Session[] = [];
+
+  try {
+    const sessionDirs = readdirSync(sessionStateDir);
+    for (const sessionId of sessionDirs) {
+      const sessionDir = join(sessionStateDir, sessionId);
+      try {
+        const dirStat = statSync(sessionDir);
+        if (!dirStat.isDirectory() || dirStat.mtime < since) continue;
+
+        // Look for session files within the subdirectory
+        const sessionFiles = readdirSync(sessionDir);
+        let found = false;
+
+        for (const file of sessionFiles) {
+          if (found) break;
+          const filePath = join(sessionDir, file);
+          try {
+            const fileStat = statSync(filePath);
+            if (!fileStat.isFile()) continue;
+            if (!file.endsWith(".json") && !file.endsWith(".jsonl") && !file.endsWith(".yaml")) continue;
+
+            const fd = openSync(filePath, "r");
+            const buf = Buffer.alloc(4096);
+            const bytesRead = readSync(fd, buf, 0, 4096, 0);
+            closeSync(fd);
+            const text = buf.toString("utf-8", 0, bytesRead);
+
+            // Look for cwd in JSON/JSONL metadata
+            for (const line of text.split("\n").slice(0, 15)) {
+              if (!line.trim()) continue;
+              try {
+                const obj = JSON.parse(line);
+                if (obj.cwd && existsSync(obj.cwd)) {
+                  sessions.push({ tool: "copilot", cwd: obj.cwd });
+                  found = true;
+                  break;
+                }
+              } catch {
+                continue;
+              }
+            }
+          } catch {
+            continue;
+          }
+        }
+      } catch {
+        continue;
+      }
+    }
+  } catch {
+    // Directory read error
+  }
+
+  return sessions;
+}
 
 async function resolveAndDeduplicate(sessions: Session[]): Promise<Repo[]> {
   // Group sessions by cwd
@@ -496,7 +559,7 @@ async function resolveAndDeduplicate(sessions: Session[]): Promise<Repo[]> {
       }
     }
 
-    const sessionCounts = { claude_code: 0, codex: 0, gemini: 0 };
+    const sessionCounts = { claude_code: 0, codex: 0, gemini: 0, copilot: 0 };
     for (const s of data.sessions) {
       sessionCounts[s.tool]++;
     }
@@ -512,8 +575,8 @@ async function resolveAndDeduplicate(sessions: Session[]): Promise<Repo[]> {
   // Sort by total sessions descending
   repos.sort(
     (a, b) =>
-      b.sessions.claude_code + b.sessions.codex + b.sessions.gemini -
-      (a.sessions.claude_code + a.sessions.codex + a.sessions.gemini)
+      b.sessions.claude_code + b.sessions.codex + b.sessions.gemini + b.sessions.copilot -
+      (a.sessions.claude_code + a.sessions.codex + a.sessions.gemini + a.sessions.copilot)
   );
 
   return repos;
@@ -530,12 +593,13 @@ async function main() {
   const ccSessions = scanClaudeCode(sinceDate);
   const codexSessions = scanCodex(sinceDate);
   const geminiSessions = scanGemini(sinceDate);
+  const copilotSessions = scanCopilot(sinceDate);
 
-  const allSessions = [...ccSessions, ...codexSessions, ...geminiSessions];
+  const allSessions = [...ccSessions, ...codexSessions, ...geminiSessions, ...copilotSessions];
 
   // Summary to stderr
   console.error(
-    `Discovered: ${ccSessions.length} CC sessions, ${codexSessions.length} Codex sessions, ${geminiSessions.length} Gemini sessions`
+    `Discovered: ${ccSessions.length} CC sessions, ${codexSessions.length} Codex sessions, ${geminiSessions.length} Gemini sessions, ${copilotSessions.length} Copilot sessions`
   );
 
   // Deduplicate
@@ -547,6 +611,7 @@ async function main() {
   const ccRepos = new Set(repos.filter((r) => r.sessions.claude_code > 0).map((r) => r.remote)).size;
   const codexRepos = new Set(repos.filter((r) => r.sessions.codex > 0).map((r) => r.remote)).size;
   const geminiRepos = new Set(repos.filter((r) => r.sessions.gemini > 0).map((r) => r.remote)).size;
+  const copilotRepos = new Set(repos.filter((r) => r.sessions.copilot > 0).map((r) => r.remote)).size;
 
   const result: DiscoveryResult = {
     window: since,
@@ -556,6 +621,7 @@ async function main() {
       claude_code: { total_sessions: ccSessions.length, repos: ccRepos },
       codex: { total_sessions: codexSessions.length, repos: codexRepos },
       gemini: { total_sessions: geminiSessions.length, repos: geminiRepos },
+      copilot: { total_sessions: copilotSessions.length, repos: copilotRepos },
     },
     total_sessions: allSessions.length,
     total_repos: repos.length,
@@ -566,15 +632,16 @@ async function main() {
   } else {
     // Summary format
     console.log(`Window: ${since} (since ${startDate})`);
-    console.log(`Sessions: ${allSessions.length} total (CC: ${ccSessions.length}, Codex: ${codexSessions.length}, Gemini: ${geminiSessions.length})`);
+    console.log(`Sessions: ${allSessions.length} total (CC: ${ccSessions.length}, Codex: ${codexSessions.length}, Gemini: ${geminiSessions.length}, Copilot: ${copilotSessions.length})`);
     console.log(`Repos: ${repos.length} unique`);
     console.log("");
     for (const repo of repos) {
-      const total = repo.sessions.claude_code + repo.sessions.codex + repo.sessions.gemini;
+      const total = repo.sessions.claude_code + repo.sessions.codex + repo.sessions.gemini + repo.sessions.copilot;
       const tools = [];
       if (repo.sessions.claude_code > 0) tools.push(`CC:${repo.sessions.claude_code}`);
       if (repo.sessions.codex > 0) tools.push(`Codex:${repo.sessions.codex}`);
       if (repo.sessions.gemini > 0) tools.push(`Gemini:${repo.sessions.gemini}`);
+      if (repo.sessions.copilot > 0) tools.push(`Copilot:${repo.sessions.copilot}`);
       console.log(`  ${repo.name} (${total} sessions) — ${tools.join(", ")}`);
       console.log(`    Remote: ${repo.remote}`);
       console.log(`    Paths: ${repo.paths.join(", ")}`);