flow-wizard
Pass
Audited by Gen Agent Trust Hub on May 26, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXPOSURE]: The skill reads local filesystem data, including project configuration files and resource definitions (services, events, domains). This is necessary for the skill's primary function of cross-referencing existing architecture. No sensitive data beyond project metadata is accessed, and no network exfiltration was detected.
- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection because it ingests data from external project files (e.g., eventcatalog.config.js, index.mdx) and incorporates that data into newly generated documentation. However, the risk is characterized as low because the skill lacks the capabilities to perform dangerous operations like credential theft or remote code execution using that data.
- Ingestion points: Reads configuration and resource files from the local filesystem to build an inventory (SKILL.md, Step 1).
- Boundary markers: Not explicitly defined; the skill processes data based on structured file paths and YAML frontmatter keys.
- Capability inventory: Filesystem read/write operations and resource discovery via the EventCatalog MCP server.
- Sanitization: Data is mapped to specific documentation fields (IDs, names, versions) rather than being executed as instructions.
Audit Metadata