lean-update
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches skill inventories, updates, and documentation from the author's GitHub repository at
github.com/aussadawut-dev/lean-skills. - [INDIRECT_PROMPT_INJECTION]: The skill's workflow creates a surface for indirect prompt injection by ingesting and interpreting external text (metadata and documentation) during the update process.
- Ingestion points:
SKILL.md(Step 2 and Step 5) fetches and reads upstream documentation and frontmatter from external repository content. - Boundary markers: No explicit delimiters or instructions to ignore embedded natural language commands are implemented when processing the external content.
- Capability inventory: The skill possesses file system write access (for backups and installation) and network access (to fetch updates).
- Sanitization: The skill performs structural validation (frontmatter keys, file completeness) in Step 5, but there is no evidence of sanitization for malicious natural language instructions hidden within the documentation or descriptions.
Audit Metadata