doc-coauthoring
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to gather and process context from external, potentially untrusted sources, which could contain instructions meant to influence the agent's behavior.
- Ingestion points: The skill explicitly requests and processes data from shared document links, user-provided files, and integrated messaging platforms like Slack and Teams during Stage 1 (SKILL.md).
- Boundary markers: The workflow lacks specific delimiters or 'ignore embedded instructions' warnings when handling the gathered external context.
- Capability inventory: The skill utilizes tools for file creation and modification (
create_file,str_replace) and has the capability to trigger sub-agents for document testing. - Sanitization: There is no evidence of sanitization, escaping, or schema validation for the information retrieved from external links or integrated third-party tools.
Audit Metadata