hooks-permission-request-hook

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill assembles and writes shell scripts (permission-request.sh and test-permission-hook.sh) to the local filesystem based on the detected project environment using the Write tool.\n- [COMMAND_EXECUTION]: It uses chmod to modify file permissions and executes the generated test harness script to verify the hook logic immediately after creation.\n- [COMMAND_EXECUTION]: The skill modifies the agent core configuration file (.claude/settings.json) to register a new PermissionRequest hook, which automates future tool execution decisions.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via the technology stack detection process, which ingests data from local project filenames and configuration files (ingestion points: project filenames like package.json and Cargo.toml, and the content of .claude/settings.json).\n- [PROMPT_INJECTION]: Boundary markers and sanitization are absent in the generated script template, which relies on regex matching of tool inputs without additional escaping or validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 02:17 AM