ecosystem-alignment

Ecosystem Alignment

Activates when checking upstream compatibility, auditing spec compliance, or aligning with platform updates.

When to Trigger

• Claude Code version update detected
• User mentions "alignment", "upstream", "spec compliance", "version check"
• New agentskills.io spec changes published
• GSD upstream (get-shit-done) version update

Checklist

1. Version Check

claude --version                    # Claude Code
npm list get-shit-done 2>/dev/null  # GSD upstream

2. Skill Description Compliance

for f in .claude/skills/*/SKILL.md; do
  desc=$(sed -n 's/^description: //p' "$f" | head -1)
  len=${#desc}
  [ $len -gt 250 ] && echo "OVER ($len): $f"
done

agentskills.io enforces 250-character maximum.

3. Hook System Audit

• Check all hook files exist (no dead references)
• Verify CJS syntax (not ESM import in .js files)
• Test each hook executes without error
• Check for new hook types in platform (PostCompact, FileChanged, PermissionDenied, etc.)

4. Feature Gap Analysis

• Compare our skill/agent/team patterns against platform capabilities
• Identify platform features we're not using
• Identify our extensions that go beyond the platform
• Document alignment opportunities

5. Binary String Analysis

strings $(which claude) | grep -oP '"(skill|hook|agent|team|memory|worktree)[^"]*"' | sort -u

Identify new internal patterns from the installed binary.

Output

Produce an alignment report documenting:

• Current versions (Claude Code, GSD, agentskills.io)
• Spec compliance status (all skills under 250 chars?)
• Hook health (dead refs, ESM issues, missing types)
• Feature gaps (what we should adopt, what we extend beyond)
• Action items ranked by priority