The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes documentation from an external source (Context7), which creates a potential surface for indirect prompt injection if the source content is compromised or contains malicious instructions.
Ingestion points: Documentation text and code snippets are fetched from the external Context7 MCP service via the query-docs tool.
Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings for the data returned from the documentation service.
Capability inventory: The skill does not directly implement subprocess or file-system access, though the agent environment running the skill typically possesses these capabilities.
Sanitization: Absent. There are no instructions for sanitizing or validating the content returned from the external source before it is used to formulate an answer.
[DATA_EXFILTRATION]: The skill transmits user queries to the Context7 MCP service to resolve library IDs and fetch documentation.
Evidence: The instructions guide the agent to call resolve-library-id and query-docs with the user's full question to improve relevance.
Mitigation: The skill proactively mitigates data exposure risks by instructing the agent to redact API keys, passwords, tokens, and other secrets from any query before sending it to the external service.

documentation-lookup