The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill describes operations for fetching and processing user-generated content via the Comment object, which contains untrusted strings.
Ingestion points: Untrusted data enters the agent context through the GET /api/v1/comments endpoint, which returns a list of user-submitted comments.
Boundary markers: The skill body does not specify the use of delimiters or instructions for the agent to ignore potentially malicious embedded content within the comment or commentHTML fields.
Capability inventory: The skill facilitates API interactions including GET, POST, PATCH, and DELETE operations for various resources like comments, users, and badges, but does not specify shell execution or file system access.
Sanitization: While the documentation notes that commentHTML is "rendered", it does not provide guidelines for the agent to sanitize or escape this content before further processing.
[EXTERNAL_DOWNLOADS]: The skill provides instructions for installing dependencies from external sources.
The Python SDK is installed directly from the vendor's repository: pip install "fastcomments[client] @ git+https://github.com/fastcomments/fastcomments-python.git@v3.1.0".
Other language SDKs are listed from standard registries (NPM, Cargo, etc.) or vendor repositories (GitHub, Repsy).

fastcomments-api