The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes npx to fetch and execute the @claude-flow/cli package from the NPM registry at runtime. It specifically requests the @latest version, which introduces a dependency on an external source that can change without notice.
[REMOTE_CODE_EXECUTION]: By downloading and running an external CLI tool via npx during its operation, the skill performs remote code execution.
[COMMAND_EXECUTION]: The skill invokes bash commands to trigger worker hooks and scoped documentation generation via the downloaded CLI.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and compares existing code against documentation to detect drift. Malicious instructions embedded in the analyzed code could potentially influence the agent's output or tool usage.
Ingestion points: Local source code files located at the path provided in the --target argument.
Boundary markers: The instructions do not define clear delimiters or "ignore instructions" guards for the ingested code content.
Capability inventory: The skill has access to file writing (Write), shell execution (Bash/npx), and scheduled task creation (CronCreate).
Sanitization: There is no evidence of filtering or sanitization performed on the content read from the target code files.

doc-gen