speak-rate-limits

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues were identified. The skill provides reference implementations for rate limiting, exponential backoff, and batch assessment queues using standard programming patterns.
  • [INDIRECT_PROMPT_INJECTION]: The skill establishes an integration surface by processing external data (audio file paths and target text) for API consumption.
  • Ingestion points: External data enters the agent context via the Recording interface and batchAssess function parameters in SKILL.md and references/limiter-implementation.md.
  • Boundary markers: No specific delimiters or boundary warnings are present in the code templates.
  • Capability inventory: The skill configuration specifies access to Bash, Write, and Edit tools.
  • Sanitization: No sanitization of input text is provided in the reference implementation, as it is intended as a logical template for rate limiting rather than a full input validation suite.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:48 AM
Security Audit — agent-trust-hub — speak-rate-limits