The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes standard development commands including git diff to identify code changes and grep to search for existing utility functions. It also runs a benign echo command (echo simplify-guard:complete) to signal completion to the platform.
[PROMPT_INJECTION]: The skill processes untrusted data from local code changes, which presents a theoretical surface for indirect prompt injection.
Ingestion points: Unified diff output from the local repository (SKILL.md).
Boundary markers: Absent; the instructions do not specify the use of delimiters when passing code content to the analysis agents.
Capability inventory: The skill has the ability to modify local files (applying fixes) and execute shell commands.
Sanitization: Absent; the code diff is processed directly for review findings.
[SAFE]: The skill's behavior is entirely consistent with its documented purpose. No evidence of data exfiltration, hardcoded credentials, or malicious persistence mechanisms was found.

simplify