The Agent Skills Directory

[DATA_EXFILTRATION]: The skill transmits user-provided queries to an external third-party service (Context7 MCP) to resolve library IDs and fetch documentation.
Evidence: The instructions in SKILL.md require the agent to send the "user's full question" to the resolve-library-id and query-docs tools.
Mitigation: The skill includes a specific "No sensitive data" best practice requiring the agent to redact API keys, passwords, tokens, and other secrets before passing the query to the external tools.
[PROMPT_INJECTION]: The skill processes data from external documentation providers, creating a surface for indirect prompt injection attacks.
Ingestion points: Untrusted documentation content enters the agent's context through the output of the query-docs tool.
Boundary markers: Absent; the instructions do not specify the use of delimiters or provide warnings to the agent to ignore instructions embedded within the fetched documentation.
Capability inventory: While the skill itself only references documentation tools, it is designed to run within an agentic environment (e.g., Gemini CLI, Cursor) that typically has broader capabilities such as file system access and shell command execution.
Sanitization: Absent; the skill does not define methods for validating or filtering the content returned from the external documentation service.

documentation-lookup