fathom-performance-tuning

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of external meeting transcripts, which are untrusted data sources that could contain malicious instructions designed to influence the agent's behavior.
  • Ingestion points: The skill uses the Fathom API to fetch transcript payloads in SKILL.md (e.g., client.getTranscript(id)).
  • Boundary markers: While the instructions mention "data/consent guardrails," there are no explicit delimiters or prompt boundary markers defined in the provided code snippets to isolate external content.
  • Capability inventory: The skill is configured with Read, Write, and Edit tool permissions in the frontmatter, allowing the agent to modify the file system based on processed data.
  • Sanitization: No specific sanitization, filtering, or validation logic is present in the TypeScript snippets to verify the content of the transcripts before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:48 AM
Security Audit — agent-trust-hub — fathom-performance-tuning