write
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text and repository documents which represent an attack surface for instructions embedded in data.
- Ingestion points: User-provided prose and repository files read via agent instructions in SKILL.md.
- Boundary markers: No explicit delimiters or warnings are used to isolate user data.
- Capability inventory: The agent can perform repository edits and execute local shell commands.
- Sanitization: Prose input is not filtered for malicious instructions.
- [COMMAND_EXECUTION]: The agent is instructed to run a local script for punctuation verification.
- Evidence: SKILL.md contains instructions to execute scripts/check-punctuation.sh.
- Note: The command is restricted to a local script provided in the distribution and performs character-level checks without network access.
Audit Metadata