skills/security-phoenix-demo/security-skills-claude-code/opengrep-rule-generator-research/Gen Agent Trust Hub
opengrep-rule-generator-research
Pass
Audited by Gen Agent Trust Hub on Apr 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches vulnerability context and exploit examples from well-known security resources, including MITRE (cwe.mitre.org) and OWASP (owasp.org), to inform its rule generation process.
- [COMMAND_EXECUTION]: The workflow includes a recommendation to validate generated YAML rules and test files using the 'opengrep scan' command-line tool on the local environment.
- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface due to its research-driven design.
- Ingestion points: External research data is gathered from the web via 'WebSearch' and 'WebFetch' tools, including CVE details and exploit examples.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present to isolate the researched web content from the agent's instructions.
- Capability inventory: The skill possesses the capability to write YAML rule files to the local disk and execute shell commands ('opengrep scan') as described in the 'SKILL.md' process flow.
- Sanitization: The instructions do not define any sanitization, escaping, or validation logic for the content retrieved from external websites before the agent processes it to generate rules.
Audit Metadata