bradygaster · sytone · Apr 3, 2026
diff --git a/.copilot/skills/agent-collaboration/SKILL.md b/.copilot/skills/agent-collaboration/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "agent-collaboration"
 description: "Standard collaboration patterns for all squad agents — worktree awareness, decisions, cross-agent communication"
-domain: "team-workflow"
-confidence: "high"
-source: "extracted from charter boilerplate — identical content in 18+ agent charters"
+license: "MIT"
+metadata:
+  domain: "team-workflow"
+  confidence: "high"
+  source: "extracted from charter boilerplate — identical content in 18+ agent charters"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [collaboration, decisions, worktree, cross-agent, communication, squad-patterns]
+  roles: [developer, lead, architect, scribe]
 ---
 
 ## Context

diff --git a/.copilot/skills/agent-conduct/SKILL.md b/.copilot/skills/agent-conduct/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "agent-conduct"
 description: "Shared hard rules enforced across all squad agents"
-domain: "team-governance"
-confidence: "high"
-source: "reskill extraction — Product Isolation Rule and Peer Quality Check appeared in all 20 agent charters"
+license: "MIT"
+metadata:
+  domain: "team-governance"
+  confidence: "high"
+  source: "reskill extraction — Product Isolation Rule and Peer Quality Check appeared in all 20 agent charters"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [conduct, rules, product-isolation, quality-check, agent-rules, governance]
+  roles: [developer, tester, lead, architect]
 ---
 
 ## Context

diff --git a/.copilot/skills/architectural-proposals/SKILL.md b/.copilot/skills/architectural-proposals/SKILL.md
@@ -1,16 +1,15 @@
 ---
 name: "architectural-proposals"
 description: "How to write comprehensive architectural proposals that drive alignment before code is written"
-domain: "architecture, product-direction"
-confidence: "high"
-source: "earned (2026-02-21 interactive shell proposal)"
-tools:
-  - name: "view"
-    description: "Read existing codebase, prior decisions, and team context before proposing changes"
-    when: "Always read .squad/decisions.md, relevant PRDs, and current architecture docs before writing proposal"
-  - name: "create"
-    description: "Create proposal in docs/proposals/ with structured format"
-    when: "After gathering context, before any implementation work begins"
+license: "MIT"
+allowed-tools: "view create"
+metadata:
+  domain: "architecture, product-direction"
+  confidence: "high"
+  source: "earned (2026-02-21 interactive shell proposal)"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [architecture, proposal, design, alignment, documentation, planning]
+  roles: [architect, lead, developer]
 ---
 
 ## Context

diff --git a/.copilot/skills/ci-validation-gates/SKILL.md b/.copilot/skills/ci-validation-gates/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "ci-validation-gates"
 description: "Defensive CI/CD patterns: semver validation, token checks, retry logic, draft detection — earned from v0.8.22"
-domain: "ci-cd"
-confidence: "high"
-source: "extracted from Drucker and Trejo charters — earned knowledge from v0.8.22 release incident"
+license: "MIT"
+metadata:
+  domain: "ci-cd"
+  confidence: "high"
+  source: "extracted from Drucker and Trejo charters — earned knowledge from v0.8.22 release incident"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [ci, cd, semver, validation, token, npm-publish, release, retry]
+  roles: [developer, release-engineer]
 ---
 
 ## Context

diff --git a/.copilot/skills/cli-wiring/SKILL.md b/.copilot/skills/cli-wiring/SKILL.md
@@ -1,3 +1,16 @@
+---
+name: "cli-wiring"
+description: "Checklist for adding new CLI commands — create command file, add routing in cli-entry.ts, add help text"
+license: "MIT"
+metadata:
+  domain: "cli-development"
+  confidence: "high"
+  source: "extracted from issue #237 / PR #244"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [cli, command, routing, wiring, cli-entry, typescript, new-command]
+  roles: [developer]
+---
+
 # Skill: CLI Command Wiring
 
 **Bug class:** Commands implemented in `packages/squad-cli/src/cli/commands/` but never routed in `cli-entry.ts`.

diff --git a/.copilot/skills/client-compatibility/SKILL.md b/.copilot/skills/client-compatibility/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "client-compatibility"
 description: "Platform detection and adaptive spawning for CLI vs VS Code vs other surfaces"
-domain: "orchestration"
-confidence: "high"
-source: "extracted"
+license: "MIT"
+metadata:
+  domain: "orchestration"
+  confidence: "high"
+  source: "extracted"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [platform, detection, vscode, cli, spawning, compatibility, adaptive, surface]
+  roles: [architect, developer, lead]
 ---
 
 ## Context

diff --git a/.copilot/skills/distributed-mesh/SKILL.md b/.copilot/skills/distributed-mesh/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "distributed-mesh"
 description: "How to coordinate with squads on different machines using git as transport"
-domain: "distributed-coordination"
-confidence: "high"
-source: "multi-model-consensus (Opus 4.6, Sonnet 4.5, GPT-5.4)"
+license: "MIT"
+metadata:
+  domain: "distributed-coordination"
+  confidence: "high"
+  source: "multi-model-consensus (Opus 4.6, Sonnet 4.5, GPT-5.4)"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [distributed, mesh, git, coordination, multi-machine, zones, sync]
+  roles: [architect, developer, lead]
 ---
 
 ## SCOPE

diff --git a/.copilot/skills/git-workflow/SKILL.md b/.copilot/skills/git-workflow/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "git-workflow"
 description: "Squad branching model: dev-first workflow with insiders preview channel"
-domain: "version-control"
-confidence: "high"
-source: "team-decision"
+license: "MIT"
+metadata:
+  domain: "version-control"
+  confidence: "high"
+  source: "team-decision"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [git, branching, workflow, worktrees, pull-request, dev, insiders]
+  roles: [developer, lead, release-engineer]
 ---
 
 ## Context

diff --git a/.copilot/skills/github-multi-account/SKILL.md b/.copilot/skills/github-multi-account/SKILL.md
@@ -1,9 +1,15 @@
 ---
-name: github-multi-account
-description: Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically.
-confidence: high
-source: https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account
-author: tamirdresher
+name: "github-multi-account"
+description: "Detect and set up account-locked gh aliases for multi-account GitHub. The AI reads this skill, detects accounts, asks the user which is personal/work, and runs the setup automatically."
+license: "MIT"
+metadata:
+  domain: "developer-tools"
+  confidence: "high"
+  source: "https://github.com/tamirdresher/squad-skills/tree/main/plugins/github-multi-account"
+  author: "tamirdresher"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [github, multi-account, gh, aliases, personal, work, emu, authentication]
+  roles: [developer]
 ---
 
 # GitHub Multi-Account — AI-Driven Setup

diff --git a/.copilot/skills/history-hygiene/SKILL.md b/.copilot/skills/history-hygiene/SKILL.md
@@ -1,9 +1,14 @@
 ---
-name: history-hygiene
-description: Record final outcomes to history.md, not intermediate requests or reversed decisions
-domain: documentation, team-collaboration
-confidence: high
-source: earned (Kobayashi v0.6.0 incident, team intervention)
+name: "history-hygiene"
+description: "Record final outcomes to history.md, not intermediate requests or reversed decisions"
+license: "MIT"
+metadata:
+  domain: "documentation, team-collaboration"
+  confidence: "high"
+  source: "earned (Kobayashi v0.6.0 incident, team intervention)"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [history, hygiene, documentation, decisions, outcomes, history-md]
+  roles: [developer, scribe, lead]
 ---
 
 ## Context

diff --git a/.copilot/skills/init-mode/SKILL.md b/.copilot/skills/init-mode/SKILL.md
@@ -1,13 +1,15 @@
 ---
 name: "init-mode"
 description: "Team initialization flow (Phase 1 proposal + Phase 2 creation)"
-domain: "orchestration"
-confidence: "high"
-source: "extracted"
-tools:
-  - name: "ask_user"
-    description: "Confirm team roster with selectable menu"
-    when: "Phase 1 proposal — requires explicit user confirmation"
+license: "MIT"
+allowed-tools: "ask_user"
+metadata:
+  domain: "orchestration"
+  confidence: "high"
+  source: "extracted"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [init, initialization, team-setup, onboarding, roster, proposal]
+  roles: [architect, lead, developer]
 ---
 
 ## Context

diff --git a/.copilot/skills/model-selection/SKILL.md b/.copilot/skills/model-selection/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "model-selection"
 description: "Per-agent model selection with 4-layer hierarchy and fallback chains"
-domain: "orchestration"
-confidence: "high"
-source: "extracted"
+license: "MIT"
+metadata:
+  domain: "orchestration"
+  confidence: "high"
+  source: "extracted"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [model, selection, hierarchy, fallback, spawning, cost, haiku, sonnet, opus]
+  roles: [architect, lead, developer]
 ---
 
 ## Context

diff --git a/.copilot/skills/release-process/SKILL.md b/.copilot/skills/release-process/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "release-process"
 description: "Step-by-step release checklist for Squad — prevents v0.8.22-style disasters"
-domain: "release-management"
-confidence: "high"
-source: "team-decision"
+license: "MIT"
+metadata:
+  domain: "release-management"
+  confidence: "high"
+  source: "team-decision"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [release, publish, npm, semver, checklist, deployment, versioning]
+  roles: [developer, release-engineer, lead]
 ---
 
 ## Context

diff --git a/.copilot/skills/reskill/SKILL.md b/.copilot/skills/reskill/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "reskill"
 description: "Team-wide charter and history optimization through skill extraction"
-domain: "team-optimization"
-confidence: "high"
-source: "manual — Brady directive to reduce per-agent context overhead"
+license: "MIT"
+metadata:
+  domain: "team-optimization"
+  confidence: "high"
+  source: "manual — Brady directive to reduce per-agent context overhead"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [reskill, optimize, charter, context, extraction, skill, bloat]
+  roles: [architect, lead, scribe]
 ---
 
 ## Context

diff --git a/.copilot/skills/reviewer-protocol/SKILL.md b/.copilot/skills/reviewer-protocol/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "reviewer-protocol"
 description: "Reviewer rejection workflow and strict lockout semantics"
-domain: "orchestration"
-confidence: "high"
-source: "extracted"
+license: "MIT"
+metadata:
+  domain: "orchestration"
+  confidence: "high"
+  source: "extracted"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [review, rejection, lockout, protocol, approval, quality-gate]
+  roles: [developer, tester, lead, architect]
 ---
 
 ## Context

diff --git a/.copilot/skills/secret-handling/SKILL.md b/.copilot/skills/secret-handling/SKILL.md
@@ -1,9 +1,14 @@
 ---
-name: secret-handling
-description: Never read .env files or write secrets to .squad/ committed files
-domain: security, file-operations, team-collaboration
-confidence: high
-source: earned (issue #267 — credential leak incident)
+name: "secret-handling"
+description: "Never read .env files or write secrets to .squad/ committed files"
+license: "MIT"
+metadata:
+  domain: "security, file-operations, team-collaboration"
+  confidence: "high"
+  source: "earned (issue #267 — credential leak incident)"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [secrets, security, env, credentials, leak, committed-files, scribe]
+  roles: [developer, scribe, lead]
 ---
 
 ## Context

diff --git a/.copilot/skills/squad-conventions/SKILL.md b/.copilot/skills/squad-conventions/SKILL.md
@@ -1,9 +1,14 @@
 ---
 name: "squad-conventions"
 description: "Core conventions and patterns used in the Squad codebase"
-domain: "project-conventions"
-confidence: "high"
-source: "manual"
+license: "MIT"
+metadata:
+  domain: "project-conventions"
+  confidence: "high"
+  source: "manual"
+  compatibility: "GitHub Copilot CLI, VS Code Copilot Chat"
+  triggers: [conventions, patterns, codebase, squad, nodejs, zero-dependencies, windows]
+  roles: [developer, lead, architect]
 ---
 
 ## Context

diff --git a/.squad/agents/span/charter.md b/.squad/agents/span/charter.md
@@ -0,0 +1,38 @@
+# SPAN — Skill Curator
+
+> Owns skill quality, schema compliance, eval coverage, and triggering accuracy.
+
+## Role
+
+SPAN manages the lifecycle of skills across all locations (`.squad/skills/`, `.copilot/skills/`, `templates/skills/`). When a skill is added or updated, SPAN validates it against the agentskills.io specification, runs trigger evals, and gates the change on pass rates.
+
+## Responsibilities
+
+- **Schema compliance** — Validate SKILL.md frontmatter against the agentskills.io spec (`name`, `description`, `license`, `metadata`)
+- **Description optimization** — Review and improve skill descriptions using imperative phrasing, user-intent focus, and near-miss testing per agentskills.io guidance
+- **Eval coverage** — Ensure every skill has eval fixtures (min 5 positive, 3 negative, 2 edge cases)
+- **Two-phase eval execution** — Run Phase 1 (keyword, `run-evals.mjs`) and Phase 2 (LLM, `run-llm-evals.mjs`) evals
+- **Domain overlap detection** — Flag skills with >50% description keyword overlap for merge consideration
+- **Progressive disclosure** — Ensure SKILL.md stays under 500 lines, deep content in `references/`
+- **Gate skill PRs** — Block merges when eval pass rate drops below 80% (Phase 1) or trigger rate drops below 0.5 (Phase 2)
+
+## Hard Rules
+
+1. **Every skill MUST have an eval fixture** — no exceptions
+2. **Description changes require eval re-run** — never change a description without verifying trigger quality
+3. **Never optimize descriptions against validation set** — use train/validation split (60/40)
+
+## Tools
+
+- `node .squad/skills/evals/validate-schema.mjs` — Schema compliance check
+- `node .squad/skills/evals/run-evals.mjs` — Phase 1 keyword eval
+- `node .squad/skills/evals/run-llm-evals.mjs` — Phase 2 LLM eval
+
+## References
+
+- [agentskills.io specification](https://agentskills.io/specification)
+- [Optimizing descriptions](https://agentskills.io/skill-creation/optimizing-descriptions)
+- [Evaluating skills](https://agentskills.io/skill-creation/evaluating-skills)
+- [Best practices](https://agentskills.io/skill-creation/best-practices)
+- `.squad/skills/CONTRIBUTING.md` — Skill contribution workflow
+- `.squad/templates/skill-review-checklist.md` — Review checklist
diff --git a/.squad/agents/span/history.md b/.squad/agents/span/history.md
@@ -0,0 +1,23 @@
+# SPAN — History
+
+## Project Context
+- **Project:** Squad — the programmable multi-agent runtime for GitHub Copilot
+- **Stack:** TypeScript (strict, ESM), Node.js ≥20, Vitest, esbuild
+- **Owner:** Brady
+- **Universe:** Apollo 13 / NASA Mission Control
+
+## Learnings
+
+### Skill Landscape (2026-04-03)
+- 34 skills across 3 canonical locations: `.squad/skills/` (14), `.copilot/skills/` (17), `templates/skills/` (3)
+- Skills synced to packages via `scripts/sync-skill-templates.mjs` (source: `.squad/skills/`)
+- SDK scans `.copilot/skills/` first (primary), falls back to `.squad/skills/` (legacy)
+- SDK matching: `triggers` array (case-insensitive substring, +0.5/hit capped at 0.7) + `roles` affinity (+0.3)
+
+### Schema Standard
+- agentskills.io spec: `name`, `description`, `license` as top-level; `domain`, `confidence`, `source`, `triggers`, `roles`, `compatibility` in `metadata`
+- Spec reference: https://agentskills.io/specification
+
+### Eval Baseline (2026-04-03)
+- Phase 1 (keyword): 88.9% pass rate (304/342 test cases, 31 fixtures)
+- Phase 2 (LLM): pending initial baseline run