init-devin-review

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is designed to assist in generating markdown-based configuration files for the Devin AI agent. It contains no malicious code, obfuscated payloads, or unauthorized data exfiltration mechanisms.\n- [COMMAND_EXECUTION]: The skill references the npx devin-review command in its documentation and evaluation sets. This is the official CLI tool for the Devin service, used to verify configuration and trigger automated PR reviews. This is a neutral, service-specific finding as per established trust rules.\n- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it is instructed to read existing repository instruction files to ground its generation. \n
  • Ingestion points: The skill reads files such as REVIEW.md, AGENTS.md, CLAUDE.md, CONTRIBUTING.md, .cursorrules, .windsurfrules, *.rules, and *.mdc (SKILL.md).\n
  • Boundary markers: There are no explicit delimiters or specific warnings to ignore embedded instructions when reading these external files.\n
  • Capability inventory: The skill outputs markdown configuration files and suggests using the devin-review CLI tool.\n
  • Sanitization: No specific sanitization or filtering of the content within the ingested instruction files is mentioned in the workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 02:47 PM