What is skill-refinement?

A review tool for measuring how well other skills work over time. It examines task records, unresolved questions, costs, failures, repeated work, and changes in confidence.

Use it to compare skill performance, find high-friction or efficient skills, and propose changes based on recorded results.

It helps identify skills that often fail, leave important questions unanswered, or use more effort than expected. That gives maintainers evidence for improving those skills.

skill-refinement — Skill for Claude Code and Codex by Adelie-Squad: A review tool for measuring how well other skills…

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add skills/adelie-squad/solosquad/skill-refinement

Any agent

npx skills add Adelie-Squad/solosquad --skill skill-refinement

Clone the repo

git clone --depth 1 https://github.com/Adelie-Squad/solosquad

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/adelie-squad/solosquad/skill-refinement.svg)](https://agentmods.dev/skills/adelie-squad/solosquad/skill-refinement)

Your own site

<a href="https://agentmods.dev/skills/adelie-squad/solosquad/skill-refinement"><img src="https://agentmods.dev/badge/skills/adelie-squad/solosquad/skill-refinement.svg" alt="Measured on agentmods" height="20"></a>

Per session 39 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 591 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00039	$0.00591
Opus 5	$0.00019	$0.00296
Sonnet 5	$0.00008	$0.00118
Haiku 4.5	$0.00004	$0.00059

Measured 4d ago against content hash aab1edda35a2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

skill-refinement scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/skill-refinement/SKILL.md · 78 lines

What it actually says

Chief 자가학습 루프의 일부. 어떤 skill 이 잘 작동하고, 어떤 skill 이 개선 필요한지 평가.

입력

평가 대상 period
<org>/memory/ledger/<task-id>.jsonl — skill 호출 trace
<org>/memory/open-questions/<id>.json — 미해결/blocking 비율
<org>/memory/agent-costs.jsonl — skill 별 cost

평가 차원

per_skill_metrics:
  skill_name: <name>
  invocation_count: <N>
  avg_duration_ms: <N>
  avg_cost_usd: <N>
  hard_gate_failure_rate: <%>      # gate 통과 못 한 비율
  open_question_rate: <%>           # 호출당 발생한 open_question 수
  rework_rate: <%>                  # 같은 task 에서 재호출된 비율
  confidence_delta_avg: <+/->       # 호출 전후 confidence 변화

Failure Patterns

patterns:
  - skill: problem-definition
    issue: "TDCC P4 단계 unknown 필드 too frequent (62%)"
    root_cause_hypothesis: "archive context 너무 짧음 (avg N=4 entries)"
    fix_candidates:
      - "기본 archive window 확대 (4→8 entries)"
      - "TDCC P4 reasoning step skip threshold 도입"

출력

period: "..."
top_3_high_friction_skills: [...]
top_3_efficient_skills: [...]
refinement_proposals:
  - skill: "..."
    proposal: "..."
    expected_impact: "..."
    effort: small | medium | large
    approval_required: founder | chief_auto

Anti-Sycophancy

❌ "skill 들이 잘 작동합니다"
✅ "problem-definition 의 P4 단계가 62% 빈도로 unknown 필드 발생. archive context 확대로 개선 가능 (예상 -30%)."

Reference

v1.1 PRD §5.2 RETROSPECT + §1.1 #7 (자가학습 메커니즘)

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

4d ago First seen · 78 lines · 39 tokens per session scan A aab1edda35a2

Subscribe to this mod's changes

skill-refinement is a skill published in the GitHub repository Adelie-Squad/solosquad (19 stars, last pushed 14d ago), licensed MIT. It adds 39 tokens to every session and 591 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.