humanize-readme
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the user's
README.mdand project metadata files to perform a rewrite, which creates a vulnerability where instructions hidden in those files could override the agent's intended behavior. - Ingestion points: The skill reads the README file and project manifests (
package.json,pyproject.toml) using thecatcommand inSKILL.md. - Boundary markers: The instructions do not define delimiters or provide clear "ignore embedded instructions" warnings for the agent when processing the content of these files.
- Capability inventory: The skill is limited to reading files (
cat) and directory listing (ls). It does not request network access or file-writing capabilities. - Sanitization: There are no requirements for sanitizing or escaping the content read from the filesystem before the agent rewrites it.
Audit Metadata