ln-161-skill-creator
Pass
Audited by Gen Agent Trust Hub on Apr 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches reference files and procedural rules from the author's GitHub repository (
levnikolaevich/claude-code-skills) usingWebFetchif local files are missing. This is a standard fallback mechanism for accessing vendor-provided configuration. - [PROMPT_INJECTION]: Vulnerable to indirect prompt injection through documentation processing. The skill extracts instructions from external markdown files to create executable commands, which could be exploited if an attacker can modify project documentation.
- Ingestion points: Documentation files including
docs/**/*.md,tests/README.md,README.md, andCONTRIBUTING.md. - Boundary markers: None. The skill reads markdown sections directly based on headers and line ranges.
- Capability inventory: File system read/write access and the ability to generate commands that execute bash blocks or invoke other skills.
- Sanitization: No sanitization or validation of the input content is performed beyond structural markdown transformation.
Audit Metadata