attio-core-workflow-b
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill workflow involves querying and processing data from external Attio CRM endpoints, which serves as a potential ingestion point for untrusted content.
- Ingestion points: Data enters the agent context via API responses for list entries, record attributes, notes, and tasks as specified in the instructions of
SKILL.md. - Boundary markers: The skill defines 'Approval Boundaries' in
SKILL.mdthat require human review for bulk changes or destructive removals, mitigating accidental execution of injected commands. - Capability inventory: The skill is granted access to the
WriteandEdittools, which can be used to modify the local filesystem based on data retrieved from the API. - Sanitization: The skill does not define specific sanitization or filtering logic for the data ingested from the external API.
- [EXTERNAL_DOWNLOADS]: The skill retrieves API guidance and documentation from Attio's official domain (
docs.attio.com), which is a well-known service.
Audit Metadata