The Agent Skills Directory

[PROMPT_INJECTION] (LOW): The skill documentation identifies workflows that ingest data from untrusted external sources, exposing the agent to potential indirect prompt injection attacks.\n- Ingestion points: External tools mapped via placeholders such as ~~chat, ~~user feedback, and ~~meeting transcription in REFERENCE.md.\n- Boundary markers: No specific boundary markers or delimiters for untrusted content are defined in the documentation.\n- Capability inventory: No executable scripts were provided in the skill files, but the reference describes interactions with external MCP servers for various tools.\n- Sanitization: The documentation does not describe any sanitization or validation protocols for data pulled from external connectors.

metrics-tracking