fireflies-data-handling
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill references official Fireflies.ai documentation and API endpoints (https://api.fireflies.ai/graphql) for legitimate integration and governance purposes.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external meeting transcripts and AskFred outputs, which represents a potential injection surface. However, the skill explicitly mandates deterministic redaction, encryption, and strict approval boundaries to mitigate risks. 1. Ingestion points: Meeting transcripts, summaries, and AskFred outputs mentioned in SKILL.md. 2. Boundary markers: Explicit instructions in SKILL.md to require approval before export or model processing. 3. Capability inventory: File system access tools (Read, Glob, Grep, Write, Edit) specified in allowed-tools frontmatter. 4. Sanitization: Mandatory deterministic redaction, encryption of stores, and restricted export paths defined in the instructions.
Audit Metadata