papercuts
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill follows security best practices by explicitly instructing the agent not to log secrets, credentials, personal data, or sensitive customer payloads into the papercuts file.
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it is designed to read, review, and resolve entries from the
.agents/PAPERCUTS.mdfile, which could be modified by external actors or malicious contributors. - Ingestion points: The
.agents/PAPERCUTS.mdfile is read during the "Review or resolve" phase (SKILL.md). - Boundary markers: None; the skill uses a simple markdown list format without explicit delimiters or instructions to ignore embedded commands.
- Capability inventory: The skill has the capability to write to the file system and modify the repository to resolve identified issues.
- Sanitization: No sanitization or validation of the log entries is specified before the agent processes them.
Audit Metadata