Common Skill CreatorSAFE

A collection of Agent Skills Standard and Best Practice for Programming Languages, Frameworks that help our AI Agent follow best practies on frameworks and programming laguages

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: common-skill-creator
description: "Standardizes the creation and evaluation of high-density Agent Skills (Claude, Cursor, Windsurf). Ensures skills achieve high Activation (specificity/completeness) and Implementation (conciseness/actionability) scores. Use when: writing or auditing SKILL.md, improving trigger accuracy, or refactoring skills to reduce redundancy and maximize token ROI."
metadata:
  triggers:
    files:
      - "SKILL.md"
      - "evals/evals.json"
    keywords:
      - create skill
      - audit skill
      - trigger rate
      - optimize description
---

# Agent Skill Creator Standard

## **Priority: P0 (CRITICAL)**

Applies to **every skill in this registry**. Maximize **Token ROI**. Every line in SKILL.md must provide specific procedural value. **Activation** (how it triggers) and **Implementation** (how it helps) primary quality metrics.

## Three-Level Loading System

- **Level 1** Frontmatter: `name` + `description` (Activation Anchor), ≤100 words.
- **Level 2** SKILL.md body: Core Rules + Workflows (Implementation Core), ≤100 lines.
- **Level 3** references/: Detailed examples, schemas, and "TESTS.md" (On-demand).

## Workflow (New or Existing Skill)
**New skill:**
1. **Research** — web-search domain best practices, checklists, and standards; extract key terms → triggers, workflows → guidelines, mistakes → anti-patterns. See [Web Search Research](references/web-search-research.md).
2. **Capture intent** — what it , when it trigger, expected output format?
3. **Draft the SKILL.md** — draft using [TEMPLATE.md](references/TEMPLATE.md)
4. **Test** — spawn parallel subagents: one with-skill, one without-skill (baseline)
5. **Evaluate** — grade assertions, review benchmark (pass rate, tokens, time)
6. **Iterate** — rewrite based on feedback, rerun into next iteration dir, repeat
7. **Optimize description** — run trigger eval queries, target ≥80% accuracy
8. **Pressure-test** — for discipline skills, capture agent excuses, red flags, and stop conditions
   **Existing skill:**
9. **Audit** — run Quality Checklist below; identify violations
10. **Snapshot** — `cp -r <skill-dir> <workspace>/skill-snapshot/` before any edits
11. **Improve SKILL.md** — fix violations, compress, move oversized content to `references/`
12. **Test** — spawn parallel subagents: one with-new-skill, one with-snapshot (baseline)
13. **Evaluate & iterate** — same as steps 4–5 above
14. **Optimize description** — re-run trigger eval if description changed
15. **Harden** — add rationalization counters where agents still fail under pressure
    See [Eval Workflow](references/eval-workflow.md) for full testing + iteration details.

## Description Quality (Activation)

- **Third-Person Voice**: Use `Standardizes...`, `Audits...`, `Encrypts...`. Avoid "I will" or "This skill helps to".
- **What + When Structure**:
- **What**: Define 5–8 specific capabilities (e.g., "Generates JWT tokens, rotates keys").
- **When**: Explicitly define triggers (e.g., "Use when user says 'rotate keys'").
- **Specificity**: Avoid vague verbs like "manage" or "handle". Use "Validate", "Inject", "Refactor", "Sanitize".
- **Trigger Hint**: Include `(triggers: *.ext, keyword)` suffix for technical skills.
## Content Quality (Implementation)

- **No Redundant Knowledge**: Do NOT explain concepts AI already knows (e.g., standard HTTP codes, common language syntax, basic SOLID definitions). Focus strictly on project-specific rules and constraints.
- **Readable Compact Style**: Size is an editorial budget, not a behavioral quality claim. Prefer readable, compact, imperative language over filler words, conversational phrasing, or speculative prose.
- **Actionability**: Examples must be copy-paste ready and executable.
- **Workflow Clarity**: Use sequential ordered lists for multi-step processes.
- **Progressive Disclosure**: Move code blocks >10 lines to `references/`.
- **Pressure Hardening**: Discipline skills must name red flags, common excuses, and exact stop/restart conditions.

## Evidence Tiers & Guardrails

- **Three Evidence Tiers**: Distinguish (1) structural checks (line counts, frontmatter schema), (2) textual transcript evidence (assertions, rationalization counters), and (3) executable outcomes (runnable verification, verifier pass/fail).
- **Rule Retirement & Ablation**: A rule is a candidate for retirement when repeated, representative, risk-appropriate evaluations (including adversarial cases) show zero regression without it. Candidate status does not authorize removing safety or approval controls; retain host and safety guardrails until verified deterministic runtime enforcement replaces them.
- **Pressure Scenarios**: For discipline skills, record rationalizations agents use to skip rules, define red flags that trigger immediate stop/restart, and score against observable evidence.

## Anti-Patterns

- **No "AI-splaining"**: Do not explain why a pattern is good unless it is a unique project constraint.
- **No Answer-Anchor Padding**: Do not insert arbitrary keywords or phrases solely to satisfy string-match tests.
- **No Vague Triggers**: Never use `src/**` or `**/*`. Keep triggers surgical.
- **No Description Bloat**: If description exceeds 100 words, move capabilities to body.
- **No Long Code Blocks**: Blocks >10 lines must be extracted to `references/`.
- **No Untested Guardrails**: Rules not validated against baseline failure are unverified speculation.

## Quality Checklist (Tessl-Aligned)

- [ ] **Activation ≥ 90%**: Description covers capabilities ("What") and triggers ("When").
- [ ] **Implementation ≥ 90%**: No general-purpose explanations; all examples executable.
- [ ] **Structural Compliance**: SKILL.md ≤ 100 lines; code blocks moved to `references/`.
- [ ] Trigger rate ≥80% on should-trigger queries.
- [ ] Guardrail skills include rationalizations, red flags, behavior eval fields, and should/should-not trigger cases.
## References

- [Skill Template](references/TEMPLATE.md) — load when starting

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.