The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or dangerous execution methods were found. The skill's functionality for reading documentation and inspecting code changes via git diff is standard for developer-oriented AI skills and limited to the project context.
[PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes external data from specification files and code repositories. 1. Ingestion points: Specification files in docs/superpowers/specs/ and docs/superpowers/plans/, as well as output from git diff HEAD. 2. Boundary markers: The skill does not implement explicit delimiters or instruction-ignore warnings for the ingested data. 3. Capability inventory: The skill utilizes file read operations and git diff commands. 4. Sanitization: No specific sanitization or escaping of the ingested content is performed before processing. Despite this surface, the strong internal constraints of the skill—which strictly forbid code generation—substantially mitigate the risk of successful exploitation.

guide-me