scope-appropriate-architecture

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill reads external codebase files such as READMEs, source code, and configuration files to determine project tiers and architectural patterns. This creates a surface for indirect prompt injection where malicious content in a repository could influence the agent's behavior. \n- Ingestion points: The skill uses Read, Glob, and Grep tools to access README.md, package.json, and other source files. \n- Boundary markers: Absent. No specific delimiters or instructions to ignore embedded content were found in the skill logic. \n- Capability inventory: The skill utilizes Read, Glob, Grep, WebFetch, and WebSearch tools for codebase analysis and external research. \n- Sanitization: Absent. The skill does not describe mechanisms to sanitize or validate the content read from files before it influences architectural recommendations. \n- [COMMAND_EXECUTION]: The skill instructions in rules/tier-detection-evidence.md provide specific shell commands (e.g., find, wc, grep, and python3) for the agent to analyze project scale and infrastructure. This involves executing commands on the user's local system based on instructions provided within the skill's reference materials.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 02:15 AM
Security Audit — agent-trust-hub — scope-appropriate-architecture