The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill fetches and displays user-generated content from DEV.to, including article titles, descriptions, full bodies, and comments. This content is untrusted and could contain hidden instructions intended to manipulate the agent's behavior.
Ingestion points: Article data and comments are ingested via API calls in scripts/devto.py (e.g., in the article and comments commands).
Boundary markers: The skill does not use delimiters or specific instructions to the agent to treat the retrieved content as untrusted data.
Capability inventory: The skill is capable of making network GET requests to the DEV.to API. It does not have capabilities for file system modification, environment variable access, or arbitrary command execution.
Sanitization: Basic regex is used to strip HTML tags from comments, but the skill does not perform any sanitization for natural language instructions or prompt injection patterns.
[DATA_EXFILTRATION]: The skill performs network GET requests to https://dev.to/api to retrieve article and tag information. These operations are directed at a well-known technology service and are necessary for the skill's primary functionality.

ez-devto