The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, including partner agreements and target account lists, which could contain adversarial instructions designed to influence the agent's behavior.
Ingestion points: Signed partner agreements, partner profiles, and target account lists (SKILL.md).
Boundary markers: Instructions do not include boundary markers or warnings to disregard instructions embedded in input data.
Capability inventory: The skill has the capability to orchestrate meetings via calendar-pipeline and generate documentation via proposal-automation based on external inputs (SKILL.md).
Sanitization: No sanitization or content filtering is specified for the ingested partner documents.

alliance-gtm