unslop
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze arbitrary codebase files, creating an attack surface for instructions embedded in data.
- Ingestion points: The agent reads project source code, configuration files (e.g.,
package.json,Dockerfile,serverless.yml), and Git history usinggit grepandgit logcommands. - Boundary markers: The instructions lack specific requirements for using delimiters or boundary warnings to prevent the agent from being influenced by instructions found within the audited files.
- Capability inventory: The skill enables the agent to perform file deletions, execute Git commits, and run repository-local tools like linters or static analyzers.
- Sanitization: There is no mention of sanitizing or filtering external content (code or comments) before the agent processes it to identify 'dead code' or 'slop'.
Audit Metadata