tech-brief

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses restricted bash tools (Bash(git:), Bash(node:)) to execute git commands for history analysis and a local node script for context resolution. These operations are appropriate for the tool's technical purpose and the execution of node is limited to a local script.
  • [PROMPT_INJECTION]: The skill aggregates and summarizes information from untrusted documentation files, source code, and git history, creating a surface for indirect prompt injection. \n
  • Ingestion points: Documentation files (tech-spec, architecture, feasibility), source code files, and git commit history. \n
  • Boundary markers: None. The workflow does not specify the use of delimiters or 'ignore' instructions for the ingested content. \n
  • Capability inventory: Includes the Write tool for saving briefs and Bash tools for technical data gathering. \n
  • Sanitization: Implements a 'Secret redaction scan' to detect and mask API keys and private keys, but does not provide sanitization for embedded instructions or malicious patterns in the processed text.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 01:44 PM
Security Audit — agent-trust-hub — tech-brief