workflow-automator
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied manual workflow descriptions which are considered untrusted data and could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The agent gathers descriptions of workflows from the user as specified in
SKILL.md(Step 1) andreferences/intake.md. - Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for the ingested data, increasing the risk that the agent may follow instructions embedded within the business descriptions.
- Capability inventory: The skill has access to a significant set of tools including
Bash,Write,Read,Glob,Grep,WebSearch, andWebFetch(defined inSKILL.mdfrontmatter). - Sanitization: No explicit sanitization, validation, or filtering of the user input is described before it is processed by the agent.
Audit Metadata