blueprint-adr-list

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill identifies a potential attack surface for indirect prompt injection by ingesting data from external sources.
  • Ingestion points: The skill reads files from the docs/adrs/ directory using commands like fd, awk, and grep to extract metadata.
  • Boundary markers: There are no explicit boundary markers or instructions to the agent to treat the content of the ADR files as untrusted or to ignore any instructions found within them.
  • Capability inventory: The skill uses shell tools (Bash) to perform file system searches and text processing.
  • Sanitization: Content extracted from the ADRs (such as Titles, Status, and Dates) is printed directly into a markdown table without escaping or sanitization, which could allow malicious content inside an ADR file to influence the agent's behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 04:16 PM
Security Audit — agent-trust-hub — blueprint-adr-list