The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data gathered from the web, which creates a surface for indirect prompt injection attacks.
Ingestion points: The skill ingests data through the sources input field and via the WebSearch and WebFetch tools as described in SKILL.md and manifest.json.
Boundary markers: There are no explicit instructions or delimiters defined to ensure the agent treats external content strictly as data rather than potential instructions.
Capability inventory: The skill utilizes WebSearch, WebFetch, and Read tools. It lacks capabilities for filesystem modification, arbitrary command execution, or external data exfiltration via POST requests.
Sanitization: The workflow does not specify sanitization, filtering, or validation steps for content retrieved from external URLs before it is processed by the agent.

claim-verification