The Agent Skills Directory

[SAFE]: No malicious patterns or security threats were identified across the instructions, metadata, or evaluation cases. The skill's behavior and requested tools (Bash, Write, Grep) are appropriate for its stated purpose of autonomous software development.
[PROMPT_INJECTION]: A technical surface for indirect prompt injection exists because the skill reads repository files to identify project constraints. However, the instructions mitigate this by requiring incremental implementation and a final verification pass, which are standard safety practices for autonomous agents.

autopilot