The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx @useclawlink/cli, which fetches and executes the ClawLink command-line interface from the NPM registry. This is the primary method for interacting with the service.
[INDIRECT_PROMPT_INJECTION]: The skill retrieves external data from Stack Exchange, which is an untrusted source. This presents a potential surface for indirect prompt injection if the agent interprets content from questions or answers as instructions.
Ingestion points: Data is ingested via shell command outputs from the Stack Exchange actions (e.g., reading questions or search results).
Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore instructions within the retrieved data.
Capability inventory: The skill allows for read and write operations on Stack Exchange via the CLI tool.
Sanitization: The instructions do not specify any sanitization or escaping of the content retrieved from the web.
[DATA_EXPOSURE]: Documentation explicitly states that credentials are stored at ~/.clawlink/credentials.json. This is a sensitive file path, though its use is standard for the functionality of this CLI tool.

stack-exchange