The Agent Skills Directory

[SAFE]: The skill's workflow and commands are consistent with its described purpose of autonomous testing. No suspicious behavior such as credential harvesting, unauthorized network access, or persistence mechanisms was detected.- [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection, which is inherent to its autonomous testing and auto-fixing capabilities. 1. Ingestion points: Test execution outputs and source code files (SKILL.md). 2. Boundary markers: No delimiters or explicit warnings are provided to the agent to ignore potentially malicious instructions within the test data. 3. Capability inventory: The skill utilizes the forge CLI to execute shell commands for running tests and applying fixes. 4. Sanitization: There is no evidence of sanitization or validation of the test outputs before they are processed by the agent.

autonomous-testing