The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to fetch the @claude-flow/cli package from the external NPM registry at runtime.
[REMOTE_CODE_EXECUTION]: The skill executes the downloaded @claude-flow/cli tool. The package originates from an organization that is not recognized as a trusted vendor, posing a risk of executing unverified code.
[COMMAND_EXECUTION]: The skill performs shell commands via npx for scoped documentation generation and utilizes CronCreate to establish persistent, recurring execution of the documentation worker.
[PROMPT_INJECTION]: The skill analyzes current project code and existing documentation to detect 'drift'. This behavior introduces a surface for indirect prompt injection if the processed files contain malicious instructions.
Ingestion points: Local project source code and markdown documentation files (SKILL.md).
Boundary markers: No delimiters or specific safety instructions are provided to the agent for handling untrusted file content during drift detection.
Capability inventory: The skill has access to Bash, Write, and specialized worker dispatch tools.
Sanitization: No sanitization or validation of the ingested code or documentation is documented.

doc-gen