scan-public-content
Pass
Audited by Gen Agent Trust Hub on Nov 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local shell script (
scan.sh) that utilizes standard Unix utilities likegrep,git, andcutto perform static analysis on the local file system. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves auditing external file content which represents an attack surface for indirect instructions. \n- Ingestion points: Content within the
skills/directory,READMEfiles,CHANGELOG.md, and plugin metadata files. \n- Boundary markers: None; the agent is instructed to read the raw string output of the scanner to identify and fix leaks. \n- Capability inventory: The skill possesses file-write capabilities (to fix leaks) and command execution via the provided bash script. \n- Sanitization: Absent; the agent relies on manual review of the data hits returned bygrep.
Audit Metadata