The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary function involves generating new test scripts and executing them at runtime to verify code robustness. This dynamic execution of generated code occurs within the local environment's test framework.
[PROMPT_INJECTION]: The skill processes external, potentially untrusted data from implementation diffs, which creates a surface for indirect prompt injection attacks where malicious instructions could be embedded in code comments or metadata.
Ingestion points: Implementation changes are ingested through git diff output as defined in break-it-prompt.md.
Boundary markers: The skill uses Markdown headers to separate implementation changes from instructions but lacks explicit delimiters that instruct the agent to ignore instructions embedded within the interpolated data.
Capability inventory: The agent has the capability to create new files and execute system commands to run the generated tests.
Sanitization: The skill does not implement technical sanitization or filtering of the ingested diff data before it is used for failure mode analysis.

adversarial-tester