doc-coauthoring
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves ingesting and processing data from external, potentially untrusted sources which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: In SKILL.md (Stage 1), the agent is instructed to fetch content from shared document links, messaging threads (Slack, Teams), and file storage (Google Drive, SharePoint) using integrated connectors.
- Boundary markers: The instructions do not define delimiters or specific markers to isolate the untrusted external content from the agent's operational instructions.
- Capability inventory: The skill utilizes powerful tools such as
create_fileandstr_replaceto modify the local filesystem, and it supports invoking sub-agents (Stage 3) to process the ingested document content. - Sanitization: The workflow lacks explicit steps for validating, escaping, or filtering the external content before it is used to generate or refine documentation sections.
Audit Metadata