The Agent Skills Directory

[SAFE]: A detailed audit of the skill instructions, markdown templates, and reference guides found no malicious patterns, obfuscation, or security risks. The skill focuses on promoting professional engineering standards.
[COMMAND_EXECUTION]: The skill directs the agent to utilize standard developer tools for legitimate purposes. This includes using the GitHub CLI (gh) for managing issues and pull requests, and executing common testing frameworks like npm test or pytest to verify code changes. These operations are within the expected scope of an engineering agent.
[DATA_EXFILTRATION]: No exfiltration or unsafe data handling patterns were identified. The skill proactively encourages security by instructing the agent to perform local audits using grep or semgrep to detect and prevent the accidental commit of secrets.
[PROMPT_INJECTION]: The skill employs role-based collaboration (e.g., 'Architect', 'Security Auditor', 'Reviewer') to structure multi-agent interactions. This methodology is used to foster constructive technical critique and does not attempt to bypass the agent's core safety guidelines or instructions.

collaborative-engineering-agent