profile-optimizer - SKILL.md Agent Skill

name: profile-optimizer description: >- Runs cue optimizer and rank commands, presents visual results, suggests removals and additions to slim the active profile. Use when user says "optimize profile", "clean up skills", "what skills am I not using", "suggest skills for this repo", or "profile review". tags: [meta, cue, profiles, optimization] category: meta version: 1.0.0 requires_mcps: [] allowed-tools: Bash

Profile Optimizer

You help users optimize their cue profile by analyzing skill usage, suggesting removals of unused skills, and recommending new skills based on the current repository.

Iron contract — never silently mutate a profile

Profile edits are user-visible and affect every future session under that profile. Never run cue skills remove, cue skills add-to-profile, or edit profile.yaml without first showing the user the exact diff and getting explicit approval. The dashboard step is read-only; the mutation step is gated. A silent removal that breaks the user's next session is the worst-possible failure mode for this skill.

When to activate

User says "optimize my profile", "clean up skills", "profile review"
User asks "what skills am I not using?"
User asks "what skills would help in this repo?"
User says "suggest skills", "recommend skills for this project"

Step 1 — Show current profile status

Run the optimizer for the active profile:

cue optimizer $(cat .cue.profile 2>/dev/null || cue current 2>/dev/null | head -1)

Present the output to the user. Highlight:

Total skills count
MCPs loaded
Required CLIs

Step 2 — Show usage ranking

Run the skill usage ranking:

cue skills rank 50

Present a summary to the user:

Top 5 most-used skills — these are essential, keep them
Skills with 0 usage — candidates for removal
Skills used <3 times — low-value, consider removing

Step 3 — Identify unused skills in the active profile

Cross-reference the profile's skills against the usage data:

# Get active profile skills
cue skills list 2>/dev/null

# Get usage data
cue skills rank 999 2>/dev/null

For each skill in the profile that has 0 usage, present it as a removal candidate.

Format your recommendation like:

🧹 Unused skills (candidates for removal):

  ⚠️  skill-name — never used in any session
  ⚠️  another-skill — never used in any session

💡 These skills add to your system prompt budget but haven't been triggered.
   Remove with: cue skills remove-from-profile <category/name>

Step 4 — Recommend skills for the current repo

Analyze the current repository to suggest relevant skills:

# Detect project type
ls package.json Cargo.toml pyproject.toml go.mod Makefile docker-compose.yml 2>/dev/null
cat package.json 2>/dev/null | head -20
ls -la .github/workflows/ 2>/dev/null

Based on what you find, suggest skills from the available pool:

cue skills available 2>/dev/null | head -40
cue skills search "<relevant-keyword>" 2>/dev/null

Format recommendations:

💡 Recommended skills for this repo:

  ✅ deployment/coolify — this repo has docker-compose.yml
  ✅ review/security-review — this repo handles auth (found JWT imports)
  ✅ stripe/stripe-webhooks — found @stripe/stripe-node in dependencies

  Add with: cue skills add-to-profile <id>

Step 5 — Present optimization summary

Show a before/after comparison:

📊 Profile Optimization Summary
─────────────────────────────────
  Current:  24 skills, 3 MCPs, 12 CLIs
  After:    18 skills, 3 MCPs, 10 CLIs  (−6 unused removed)

  Suggested additions: 3 skills relevant to this repo
  Token budget saved: ~2,400 tokens (from removing 6 unused skills)

  Apply changes? (list the exact commands)

Rules

Never auto-remove skills. Always present as suggestions and let the user decide.
Show the commands the user needs to run — don't run destructive commands yourself.
Be concise. Don't list every skill — focus on the actionable ones (unused + recommended).
Consider inheritance. Skills from core profile are shared everywhere — don't suggest removing those.
Explain WHY a skill is recommended (what in the repo triggered the suggestion).

Capture learnings

If you noticed something about this profile that future-you should know (a skill that looks unused but is actually critical during specific workflows, a recurring add-recommendation that the user keeps declining, a project type signal that suggests a different profile entirely), log it:

bin/cue-learnings log --type preference \
                     --key <profile-name>-<short-slug> \
                     --insight "<one-line description>" \
                     --confidence 1-10 \
                     --source observed

Conventions: ../skill-reviewer/references/learnings.md.