Governs the Claude Code/Codex Skill surface: reconciles source/install/catalog-visibility drift without losing cold capability. Use when users ask how many Skills load, why the catalog is truncated, want routers visible while other bundles stay cold, or need duplicate/superseded-Skill or old-cache cleanup. Not for building a marketplace manifest (use marketplace-dev) or repo audits (use marketplace-health-check).
- Stars
- 1,534
- Version
- v2
- Copies
- 0
Overview
Waxell rates it high risk: 39 findings, highest severity critical.
- Highest severity
- critical
- Findings
- 39
- What ran
- Static analysis + deep scan
A clean result means nothing alarming in what is published, not a guarantee. How the checks work
We don't republish the skill's instructions. Read them on GitHub at the reviewed commit.
Security
Security scan
Checked yesterday
Result
High risk Waxell rates it high risk: 39 findings, highest severity critical.
Checked with static analysis + deep scan: servers it launches, packages it installs at runtime, scripts piped to a shell and bundled binaries. The deep scan also read the instructions for prompt injection and data exfiltration. Nothing was run. How the checks work
Findings (39)
critical · 1
- critical
Dangerous Code Execution: Dangerous chain: exec() wrapping compile
scripts/audit_codex_skill_surface.py:510 · skillspector:AST8
high · 8
- high
Agent Snooping: Agent Config Directory Access
SKILL.md:51 · skillspector:AS1
- high
Agent Snooping: Agent Config Directory Access
references/skill-surface-governance.md:208 · skillspector:AS1
- high
Data Exfiltration: Env Variable Harvesting
scripts/audit_claude_skill_surface.py:50 · skillspector:E2
- high
Dangerous Code Execution: exec() call detected
scripts/audit_codex_skill_surface.py:510 · skillspector:AST1
- high
Agent Snooping: Agent Config Directory Access
scripts/audit_project_skill_roots.py:28 · skillspector:AS1
- high
System Prompt Leakage: Direct Prompt Extraction
tests/test_audit_codex_skill_surface.py:133 · skillspector:P6
- high
System Prompt Leakage: Direct Prompt Extraction
tests/test_audit_codex_skill_surface.py:651 · skillspector:P6
- high
System Prompt Leakage: Direct Prompt Extraction
tests/test_audit_codex_skill_surface.py:693 · skillspector:P6
medium · 30
- medium
Without declared permissions the skill's intent is opaque and cannot be validated.
SKILL.md:1 · skillspector:LP3
- medium
Rogue Agent: Session Persistence
references/skill-surface-governance.md:157 · skillspector:RA2
- medium
Rogue Agent: Session Persistence
references/skill-surface-governance.md:278 · skillspector:RA2
- medium
Dangerous Code Execution: subprocess module call
scripts/audit_claude_skill_surface.py:68 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/audit_codex_skill_surface.py:112 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
scripts/audit_codex_skill_surface.py:140 · skillspector:AST4
- medium
Dangerous Code Execution: compile() call detected
scripts/audit_codex_skill_surface.py:510 · skillspector:AST6
- medium
Dangerous Code Execution: subprocess module call
scripts/audit_codex_skill_surface.py:537 · skillspector:AST4
- medium
Agent Snooping: Skill Enumeration
scripts/audit_project_skill_roots.py:634 · skillspector:AS3
- medium
Dangerous Code Execution: subprocess module call
scripts/audit_skill_delivery.py:54 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_claude_skill_surface.py:27 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_claude_skill_surface.py:29 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_codex_skill_surface.py:169 · skillspector:AST4
- medium
Data Flow: Tainted flow: 'args' from os.environ.get (line 143, credential/environment) → subprocess.run (code execution)
tests/test_audit_codex_skill_surface.py:169 · skillspector:TT2
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_codex_skill_surface.py:852 · skillspector:AST4
- medium
Data Flow: Tainted flow: 'args' from os.environ.get (line 847, credential/environment) → subprocess.run (code execution)
tests/test_audit_codex_skill_surface.py:852 · skillspector:TT2
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_codex_skill_surface.py:856 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_project_skill_roots.py:58 · skillspector:AST4
- medium
Agent Snooping: Skill Enumeration
tests/test_audit_project_skill_roots.py:147 · skillspector:AS3
- medium
Agent Snooping: Skill Enumeration
tests/test_audit_project_skill_roots.py:182 · skillspector:AS3
- medium
Agent Snooping: Skill Enumeration
tests/test_audit_project_skill_roots.py:212 · skillspector:AS3
- medium
Agent Snooping: Skill Enumeration
tests/test_audit_project_skill_roots.py:215 · skillspector:AS3
- medium
Agent Snooping: Skill Enumeration
tests/test_audit_project_skill_roots.py:317 · skillspector:AS3
- medium
Agent Snooping: Skill Enumeration
tests/test_audit_project_skill_roots.py:198 · skillspector:AS3
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_project_skill_roots.py:345 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_skill_delivery.py:20 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_skill_delivery.py:69 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_skill_delivery.py:74 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_skill_delivery.py:107 · skillspector:AST4
- medium
Dangerous Code Execution: subprocess module call
tests/test_audit_skill_delivery.py:143 · skillspector:AST4
Versions
Version history
Each distinct content we have seen, newest first, with the verdict it got at the time.
- v22026-11-161 changed High riskgot riskier
- v12026-11-03 No issues
Files
Files
10 files, 174 KB in total. Names only; open them on GitHub.
- SKILL.md
- references/skill-surface-governance.md
- scripts/audit_claude_skill_surface.py
- scripts/audit_codex_skill_surface.py
- scripts/audit_project_skill_roots.py
- scripts/audit_skill_delivery.py
- tests/test_audit_claude_skill_surface.py
- tests/test_audit_codex_skill_surface.py
- tests/test_audit_project_skill_roots.py
- tests/test_audit_skill_delivery.py
See and control which skills your people use
Inside your organization, Waxell scans each skill and plugin before anyone installs it, approves what passes and publishes it to a private marketplace, and puts every MCP tool call behind one governed URL.
30 minutes. No commitment.
SOC 2 Type II attested · ISO/IEC 27001:2022 certified · HIPAA with a BAA · aligned to the EU AI Act and NIST AI RMF
More engineering skills and plugins
skills-searchdaymadeFinds Skills in configured local repositories before searching CCPM. Use when discovering or reusing a Skill, finding PDF/code-review… High riskEngineeringResearch1.5k
debugging-network-issuesdaymadeInvestigates network, streaming and protocol failures with falsification-first layered experiments, the general fallback for unknown root… High riskEngineeringResearch1.5k
feishu-doc-scraperdaymadeExtracts Feishu/Lark Docs, Wiki, Sheets, and Minutes (妙记) transcripts into faithful local Markdown via the lark-cli API, no LLM… High riskEngineeringResearch1.5k
ghostty-usedaymadeSnapshots local Codex/Claude CLI sessions across terminals and restores IDs in Ghostty after reboot (保存会话/重启前备份/恢复窗口), audits survivors,… High riskEngineeringAI & agents1.5k