The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes Glob and Grep tools to discover documentation and code examples within the user's workspace. These search operations are restricted to relevant project files such as agent.config.ts and existing Markdown documentation, which is necessary for the skill's primary purpose.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes untrusted user project files to generate documentation.
Ingestion points: Reads local project code and documentation files via Glob and Grep (SKILL.md).
Boundary markers: None identified.
Capability inventory: Limited to workspace search and documentation writing; no network access or privilege escalation tools are requested.
Sanitization: No explicit sanitization of ingested code content is described.
[SAFE]: No obfuscation, persistence mechanisms, or credential harvesting patterns were detected. References to @botpress/runtime are consistent with the vendor's own official SDKs.

adk-docs