Skills
Navigate
skills/prefactordev/typescript-sdk/report-agent-risk-data/Gen Agent Trust Hub

report-agent-risk-data

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze untrusted agent source code to infer and populate risk metadata, which serves as a potential surface for indirect prompt injection.
  • Ingestion points: Source code searched and read via rg commands in SKILL.md.
  • Boundary markers: None specified to delimit ingested data from agent instructions.
  • Capability inventory: File system read access via rg and write/network capabilities via the prefactor CLI and SDK.
  • Sanitization: No sanitization or validation of ingested code content is performed.
  • [COMMAND_EXECUTION]: The skill executes rg (ripgrep) to discover span types and examine logic within the codebase.
  • [SAFE]: All referenced packages such as @prefactor/core, @prefactor/langchain, and @prefactor/ai, along with the prefactor CLI, are authoritative vendor resources provided by the skill author (prefactordev) for the intended instrumentation task.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 06:19 PM