The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill retrieves documentation artifacts from official sources including developers.openai.com, learn.chatgpt.com, and code.claude.com. These downloads are managed by a local helper script that implements integrity checks and caching logic.
[COMMAND_EXECUTION]: The skill performs authorized local command execution to fulfill its documentation tasks. It runs a local bash script (scripts/fetch-doc.sh), uses rg for searching text, and invokes the codex CLI to inspect local feature states, versions, and configuration schemas.
[PROMPT_INJECTION]: The skill processes external documentation content, which presents a surface for indirect prompt injection.
Ingestion points: Documentation content and web search results are ingested into the agent context.
Boundary markers: The instructions do not define specific delimiters for separating fetched content from system prompts.
Capability inventory: The skill has access to command execution tools (codex, rg) and network capabilities.
Sanitization: Content integrity is verified by the helper script using specific string markers to ensure fetched files are valid documentation artifacts.

agents-docs