The Agent Skills Directory

[SAFE]: The skill implements a robust workflow for Go code simplification, prioritizing API stability and test suite integrity. It includes clear constraints to prevent behavioral changes and requires verification of all automated findings.
[COMMAND_EXECUTION]: The workflow incorporates standard Go development commands (go build, go test, go test -race, make lint) to ensure the correctness of applied simplifications. This is consistent with the skill's primary purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes external Go source code, which constitutes a potential ingestion point for untrusted data. Ingestion points: Go source files throughout the codebase being simplified (SKILL.md Step 2/3). Boundary markers: The skill establishes clear constraints ('Zero behavior change', 'Never break the exported API') and requires the agent to verify scan results as hypotheses (SKILL.md Hard Constraints). Capability inventory: The skill utilizes go build, go test, go test -race, make lint, and git checkout (SKILL.md Step 6). Sanitization: Safety is maintained through mandatory build and test gates, race detection, and linting to ensure code integrity and functional correctness post-simplification.

go-simplify