SKILLEMALL.ai
GitHub ↗ RU Sign in

BD openclaw-agent-skill-evolution

Gunakan saat user secara eksplisit meminta evolusi skill lewat benchmark, red-teaming, dan regresi pada task tertentu.

ClawHub Agent Skills author: Suga Nick v1.0.1 MIT-0 5 files body ≈ 5 788 tokens Open the sourceclawhub.ai↗ analyzed 2 d ago

Gunakan saat user secara eksplisit meminta evolusi skill lewat benchmark, red-teaming, dan regresi pada task tertentu.

As a process D 44/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions

ProcedureInfrastructureAI and agentstype and topics are labelled automatically from the skill text
Runs in: OpenClaw Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) Goose OpenCode
JSON
Technical rating
B
86/100
safety, quality, tests
Safety 60%
95
Quality 40%
72
Run on models
none yet
Process rating
D
44/100
Unfinished process
Result and completion w 14
0
Inputs and preconditions w 11
0
When it triggers w 12
20
the three weakest of ten parameters · all ten

How to improve

  1. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 5

✓ No critical or high findings

Medium and low: 5
  • low Risky intent intent-offensive-security SKILL.md:118
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)
    NEVER log secret. REDACT API KEY/TOKEN/PASSWORD/SECRET sebelum simpan. PII: MINIMIZE → REDACT → HASH. Trust boundary: TRUSTED / SEMI-TRUSTED / UNTRUSTED (user data, external web, plugin, third-party s
    detector
  • low Risky intent intent-offensive-security SKILL.md:133
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    Semver (MAJOR.MINOR.PATCH). Perubahan struktur = MAJOR. CHANGELOG wajib tiap rilis. Gunakan Version Candidate System: CURRENT → CANDIDATE → SANDBOX → BENCHMARK → RED TEAM → APPROVAL → DEPLOY. Jangan t
  • low Risky intent intent-offensive-security SKILL.md:180
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    → DESIGN UPGRADE → BUILD CANDIDATE → TEST → BENCHMARK → RED TEAM
  • low Risky intent intent-offensive-security SKILL.md:259
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - **Version Candidate System**: CURRENT → CANDIDATE → SANDBOX → BENCHMARK → RED TEAM → APPROVAL/AUTO-APPROVAL → DEPLOY.
  • low Risky intent intent-offensive-security SKILL.md:266
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)
    - **Security Evolution**: tiap generasi lebih kuat terhadap prompt injection, tool injection, data exfiltration, secret leak, malicious skill, dependency attack, privilege escalation, untrusted conten
    detector

Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning body-long SKILL.md body ≈ 5788 tokens (recommended < 5000); move details to references/

Process rating: all ten parameters 44/100

  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 20When it triggers. No condition that starts the skill
  • 30Running it twice. 21 mutating operations with no state check
  • 40Consistency. Frontmatter name (openclaw-agent-skill-evolution) differs from the folder (agent-skill-evolution)
  • 50Failures and branches. 0 branches, has a failure section
  • 60Tools and files. Uses tools (bash, node) that frontmatter does not declare
  • 70Execution cost. Instruction body is 5788 tokens
  • 100Steps. 139 steps
  • 100Progress reporting. Reports progress
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 45 top-level sections: this looks like several domains in one skill

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +3Description length 118: 120–800 characters recommended
  • +3Output format is not stated: the model decides each time
  • +2Single-language instructions
  • +4Structure: 47 headings
  • +3Step-by-step instructions: 139 items
  • +4Has examples (1 code blocks)
  • +1License stated

Quality base 70; lint remarks subtract, signals add up to 100. Result: 72.

External checks

ClawHub: clean
This skill is a disclosed framework for improving OpenClaw skills with testing and rollback safeguards, with no hidden installer or executable payload found.
LLM: benign (high) · VirusTotal: · 25 Aug 2026