openclaw-control-center

Warn

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the project source code from an external repository (https://github.com/TianyiDataScience/openclaw-control-center.git) that is not associated with a trusted organization or well-known service.
  • [COMMAND_EXECUTION]: Instructs the agent to execute installation and build commands, including npm install, npm run build, and npm run dev:ui. This execution path runs code downloaded from the external repository.
  • [CREDENTIALS_UNSAFE]: The skill provides instructions to read a sensitive local token from ~/.openclaw/local-token to authenticate API calls. Accessing and handling raw authentication tokens in a shared context increases the risk of credential exposure.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a dashboard for external agent activity, creating a surface for indirect prompt injection via the data it monitors.
  • Ingestion points: Processes agent task titles, execution chains, memory content, and document sources stored in ~/.openclaw and openclaw.json.
  • Boundary markers: None detected; the skill does not specify delimiters or instructions for the agent to ignore embedded commands in the monitored data.
  • Capability inventory: Includes subprocess execution (npm) and network communication via the Gateway URL.
  • Sanitization: No sanitization or validation logic is mentioned for the ingested agent data.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 12, 2026, 07:50 PM
Security Audit — agent-trust-hub — openclaw-control-center