skills/vgrss/acumen/workshop/Gen Agent Trust Hub

workshop

Pass

Audited by Gen Agent Trust Hub on Apr 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests and synthesizes data from .acumen/competitors.md and external feedback sources (support tickets, NPS, interviews) referenced in .acumen/sources.md. These data points may contain adversarial instructions designed to influence the agent's output. The instructions lack explicit boundary markers or sanitization requirements to isolate this untrusted content.
  • Ingestion points: Data loaded from .acumen/competitors.md and various user signal sources (support, feedback) configured via .acumen/sources.md.
  • Boundary markers: The instructions do not define delimiters or specific "ignore embedded instructions" warnings for the processed data.
  • Capability inventory: The skill has the capability to write to the local filesystem (creating reports in .acumen/reports/ and updating .acumen/value-chain.md) and triggers subsequent agent actions via commands like /increment, /roadmap, and /measure.
  • Sanitization: No validation, escaping, or filtering of the external data is specified before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 27, 2026, 02:29 PM