start-execution-model

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The workflow section in SKILL.md mandates the execution of fuse-ai-pilot:sniper, which appears to be a tool for linting or validating changes within the developer's environment.
  • [EXTERNAL_DOWNLOADS]: The instructions require the agent to use Exa to retrieve API documentation from the internet to ensure accuracy during code generation.
  • [PROMPT_INJECTION]: The skill establishes an attack surface for indirect prompt injection by directing the agent to ingest external data. 1. Ingestion points: External web content retrieved via the Exa search tool for API verification in SKILL.md. 2. Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the retrieved search results. 3. Capability inventory: The skill documentation describes potential for command execution (fuse-ai-pilot:sniper), file system access (require('node:fs')), and network requests (fetch). 4. Sanitization: No explicit sanitization or validation of the fetched external data is prescribed.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 04:06 PM
Security Audit — agent-trust-hub — start-execution-model