The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a utility script find-polluter.sh that executes local tests using npm test. Additionally, SKILL.md provides examples of diagnostic shell commands such as security list-keychains and codesign to verify build environment states.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data such as error logs, stack traces, and test results. While this represents a potential surface for indirect prompt injection, it is the intended primary purpose of a debugging skill and lacks exploitable privilege paths in this context.
Ingestion points: Error messages, stack traces, and test output processed in SKILL.md and root-cause-tracing.md.
Boundary markers: None explicitly defined for logs.
Capability inventory: Local file reading and execution of local test suites via npm.
Sanitization: None specified for log analysis.
[SAFE]: No malicious patterns, unauthorized persistence mechanisms, or data exfiltration attempts were identified. All included scripts and documentation focus on standard developer troubleshooting workflows.

systematic-debugging