api-patterns
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- PROMPT_INJECTION (HIGH): High susceptibility to indirect prompt injection based on the skill's capabilities and ingestion points.
- Ingestion points: The skill is designed to process untrusted project files, documentation, and source code via its allowed tools (Read, Glob, Grep) and the validation script.
- Boundary markers: Absent; the skill lacks instructions to delimit external content or to explicitly ignore instructions embedded within the processed project files.
- Capability inventory: The agent is granted
WriteandEditpermissions alongside the ability to execute a Python script (scripts/api_validator.py), creating a high-impact exploitation path if the agent is subverted. - Sanitization: Absent; no input validation or sanitization routines are defined for the content ingested from the project directory.
- COMMAND_EXECUTION (MEDIUM): The skill defines a tool that executes a local script
python scripts/api_validator.pywith a user-provided project path, which allows for subprocess execution on the filesystem.
Recommendations
- AI detected serious security threats
Audit Metadata