The Agent Skills Directory

[SAFE]: The skill consists of Markdown templates and instructional references that define a structured engineering process. It does not contain any malicious code, obfuscation, or unauthorized commands.
[SAFE]: The included validation script (scripts/validate_docs.py) is a read-only tool that uses standard Python libraries to verify document structure and metadata. It includes safety checks to prevent path traversal and does not perform network operations or execute untrusted code.
[DATA_EXFILTRATION]: No patterns of data exfiltration or hardcoded credentials were found. The skill instructions explicitly warn against including secrets or sensitive data in documentation and reproduction steps.
[INDIRECT_PROMPT_INJECTION]: The skill processes repository documentation which is a potential surface for indirect prompt injection. However, it includes robust guardrails, such as requirements for human approval of changes and verification of work against evidence, mitigating the risk.
[REMOTE_CODE_EXECUTION]: The skill does not download or execute code from remote sources. All scripts and operations are performed locally using standard, well-defined tools.

swe-lifecycle