The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a bash script find-polluter.sh designed to identify tests that create unwanted files or state. The script executes npm test on local files matching a user-defined pattern, which involves dynamic execution of local code.
[COMMAND_EXECUTION]: SKILL.md provides examples of diagnostic commands for troubleshooting multi-component systems, including security list-keychains and security find-identity. While intended for debugging signing failures, these commands access sensitive macOS security configuration data.
[PROMPT_INJECTION]: The skill defines a process for the agent to ingest and analyze untrusted external data (error messages, stack traces, and component logs). This creates a vulnerability to indirect prompt injection if malicious payloads are embedded in the analyzed logs.
Ingestion points: Error logs, stack traces, and component data flow logs (SKILL.md, Phase 1).
Boundary markers: None identified. The skill lacks instructions to delimit or ignore instructions within processed data.
Capability inventory: The skill uses file system exploration and command execution via shell scripts and npm.
Sanitization: No sanitization or validation of external log content is specified before the agent processes it for root cause analysis.

systematic-debugging