run-tests
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION] (LOW): The skill defines an indirect prompt injection surface (Category 8) by instructing the agent to ingest external file content to guide its behavior.
- Ingestion points: The agent is directed to use the
Readtool on{{cookiecutter.primary_pipeline_name}}/.claude/skills/run-tests/SKILL.md. - Boundary markers: Absent; there are no delimiters or instructions to ignore embedded commands within the ingested file.
- Capability inventory: The agent uses the content to generate shell commands (
make) and provide troubleshooting guidance. - Sanitization: Absent; the skill does not specify validation or filtering for the loaded file content.
- [COMMAND_EXECUTION] (SAFE): The skill provides
makecommands for the user to execute manually. While these allow for command execution on the host, they are presented as documentation for the user's primary task of running tests and do not involve the agent executing code directly.
Audit Metadata