utility-pm-workflow-builder
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a standard authoring utility that follows best practices for repository maintenance. It restricts its output to a gitignored staging directory (
_staging/workflows/) and explicitly states that it does not execute chains or promote its own output to canonical locations, ensuring a human-in-the-loop review process. - [INDIRECT_PROMPT_INJECTION]: The skill has a vulnerability surface for indirect prompt injection because it is instructed to ingest and analyze existing workflows and user-provided ideas to generate new content.
- Ingestion points: The skill scans the
_workflows/directory and reads user-provided problem statements or chain expressions to perform overlap analysis and sequence design. - Boundary markers: The instructions do not provide specific delimiters or ignore-instructions for the content found within the scanned
_workflows/files. - Capability inventory: The skill has the capability to read file structures and write new files to the local file system (specifically in the staging directory). It does not have network access or administrative privileges.
- Sanitization: There is no mention of sanitizing or escaping the content retrieved from the workflow files or user input before it is used to generate the final output packets.
Audit Metadata