Red Team Verificator
PassAudited by VirusTotal on May 2, 2026.
Overview
Type: OpenClaw Skill Name: red-team-verificator Version: 2.1.0 The skill bundle consists of a metadata file and a markdown instruction set (SKILL.md) designed to guide an AI agent in auditing startup business strategies. The instructions focus on logical verification, fact-checking, and identifying contradictions or 'hallucinations' in business plans, specifically regarding Russian market regulations (152-FZ) and unit economics. There is no executable code, no requests for sensitive system access, and no evidence of malicious prompt injection or data exfiltration intent.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The assistant may strongly discourage ideas even when a more balanced business analysis would leave room for uncertainty.
The skill openly instructs the agent to be highly adversarial and pessimistic. This is aligned with a red-team verifier role, but it can make the output feel more authoritative or negative than a neutral review.
«Лучше убить 10 хороших идей, чем пропустить 1 плохую»; «НЕ МЯГЧАЙ — плохая идея = убить быстро»
Use this skill as a stress-test or second opinion, not as the sole basis for investment, product, or coding decisions.
There is little technical supply-chain exposure here, but users cannot independently inspect an external project homepage or source repository from the metadata.
The publisher provenance is limited. The risk is low because the skill is instruction-only and has no installable code or dependencies.
Source: unknown; Homepage: none
If provenance matters for your workspace, verify the publisher before relying on the skill.