file-review

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands including brew, git, bun, and the file-review binary to handle installation and tool operations.
  • [EXTERNAL_DOWNLOADS]: It downloads resources from the author's own infrastructure, specifically the desplega-ai/ai-toolbox repository and a custom Homebrew tap. These are documented as vendor-owned resources.
  • [PROMPT_INJECTION]: The 'Process Comments' workflow identifies a surface for indirect prompt injection. (1) Ingestion points: The agent reads and parses HTML-style markers from user-provided files. (2) Boundary markers: Absent; there are no instructions to disregard commands or instructions embedded within the extracted feedback. (3) Capability inventory: The agent has the ability to read/write files and execute shell commands. (4) Sanitization: Absent; the skill extracts feedback text and proposes edits based on that text without prior validation or filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 04:01 AM