warp-deploy-fund-deployer

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses various command-line tools to interact with the environment and external services.
  • Uses curl to send GraphQL queries to Linear and REST requests to CoinGecko and blockchain RPC nodes.
  • Uses cat and grep to read local registry metadata from hyperlane-registry.
  • Uses cast (Foundry toolkit) to query blockchain balances and contract states.
  • Uses pnpm tsx to execute local TypeScript scripts for wallet funding located in the typescript/infra directory.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing untrusted data from an external source.
  • Ingestion points: Fetches ticket titles and descriptions from the Linear API (api.linear.app) which are authored by users and untrusted.
  • Boundary markers: No specific delimiters or instructions are provided to the agent to treat the ticket content as untrusted data or to ignore embedded instructions during parsing.
  • Capability inventory: The agent has the capability to execute shell commands, read local files, and perform wallet funding operations based on the parsed data.
  • Sanitization: There is no explicit sanitization of the ticket data before it is parsed and used to influence subsequent command parameters (like chain names or token symbols).
  • [EXTERNAL_DOWNLOADS]: The skill fetches data from several external services.
  • Connects to api.linear.app for project management data.
  • Connects to api.coingecko.com for cryptocurrency price information.
  • Connects to various blockchain RPC endpoints (retrieved from the local registry) for network state information.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 08:41 PM
Security Audit — agent-trust-hub — warp-deploy-fund-deployer