The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The skill uses Bash to execute local commands such as git, find, grep, and yq for file discovery and content analysis. These are standard operations for a documentation synchronization utility and do not exhibit malicious patterns.
[EXTERNAL_DOWNLOADS] (LOW): The skill runs npx @apidevtools/swagger-cli to validate documentation. This command downloads and executes a package from the npm registry. While @apidevtools/swagger-cli is a reputable tool, this mechanism introduces a dependency on an external repository.
[PROMPT_INJECTION] (LOW): The skill is susceptible to indirect prompt injection (Category 8) because it parses data from source code and documentation files without sanitization to perform its checks.
Ingestion points: Files matching patterns like **/routes/**/*.ts, **/api/**/*.py, and openapi.yaml.
Boundary markers: Absent; the skill reads raw file content using grep/sed.
Capability inventory: Bash, Edit, Write, and mcp__github__*.
Sanitization: Absent; the comparison logic (comm) and reporting rely on raw strings extracted from the filesystem.

api-documentation