skills/modem-dev/hunk/hunk-review/Gen Agent Trust Hub

hunk-review

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes various hunk CLI commands to manage review sessions, navigate code, and manage comments. The reload command allows swapping session contents by executing nested Hunk subcommands such as diff and show.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from diffs and comments, creating an attack surface for indirect prompt injection.
  • Ingestion points: The skill reads patch content via hunk session review and user comments via hunk session comment list (SKILL.md).
  • Boundary markers: There are no delimiters or instructions to ignore embedded commands within the ingested content.
  • Capability inventory: The agent can execute CLI commands, navigate the filesystem via paths, and modify session state through comment management.
  • Sanitization: No sanitization or validation logic is defined for the content retrieved from external sources before agent processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 09:05 AM
Security Audit — agent-trust-hub — hunk-review