occ-use

Pass

Audited by Gen Agent Trust Hub on Oct 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill executes the @ruvnet/open-claude-code package using npx. The package is sourced from the official npm registry and belongs to the skill's author ('ruvnet'). The use of a pinned version (2.1.3) follows security best practices for verifiable dependencies.- [COMMAND_EXECUTION]: The skill uses a command-line interface to perform coding tasks. It specifies security controls such as the --allowedTools flag to restrict the tool's capabilities during read operations and --max-turns to prevent excessive automated actions.- [INDIRECT_PROMPT_INJECTION]: The skill reads external project files which may contain untrusted content that could attempt to influence the agent's behavior.
  • Ingestion points: External repository files (e.g., README.md) accessed via the CLI tool (SKILL.md).
  • Boundary markers: Instructions advise the agent to inspect task-specific instructions before execution.
  • Capability inventory: The tool has file reading and writing capabilities, as well as the ability to call AI providers via API keys.
  • Sanitization: Employs tool-specific restriction flags to limit the impact of potentially malicious content in processed data.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 12, 2026, 04:46 PM