revor-company-research
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONPERSISTENCE
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a bundled Node.js script ("scripts/revor-api.mjs") to communicate with the Revor API. This is the primary mechanism for retrieving research data and is fully documented within the skill instructions.
- [PERSISTENCE]: The skill instructions direct the agent to create a persistent configuration file at "~/.config/RevorSkill/.env" to store the REVOR_API_KEY. This is a standard and acceptable practice for managing service-specific credentials for automated tools.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data which constitutes a potential attack surface:
- Ingestion points: User-provided company names, search queries, and data retrieved from the Revor API (customs records, contacts) in SKILL.md.
- Boundary markers: The instructions lack explicit delimiters to isolate external data from the agent's core instruction set.
- Capability inventory: The skill has the capability to write files to the user's home directory, execute local scripts, and make network requests to the Revor API.
- Sanitization: The bundled script "scripts/revor-api.mjs" uses JSON serialization and URL encoding when preparing API requests, reducing the risk of data-driven command injection.
- [SAFE]: The network operations are limited to the vendor's official API domain ("revor.ai"), and the skill includes explicit instructions to avoid exposing the API key in logs or chat interfaces.
Audit Metadata