bitcoin
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes interfaces for the agent to ingest and process data from the external Bitcoin network, which constitutes an untrusted source that could potentially contain malicious instructions.
- Ingestion points: The skill documents methods to retrieve blockchain and mempool data via JSON-RPC endpoints (
core-rpc.md) and REST API endpoints (features-rest-zmq.md), including transaction scripts and metadata. - Boundary markers: The documentation does not specify the use of delimiters or specific safety instructions to prevent the agent from interpreting data within transaction fields as commands.
- Capability inventory: The agent is provided with instructions to use powerful capabilities, such as sending Bitcoin (
sendtoaddress), managing encrypted wallet files (features-wallets.md), and performing filesystem operations (backupwallet). - Sanitization: There is no mention of sanitization or validation protocols for the external data being processed by the agent.
- [COMMAND_EXECUTION]: The skill details the use of administrative and financial command-line tools that interact directly with the host system and digital assets.
- Evidence: The reference files (
core-overview.md,core-cli.md) describe the execution ofbitcoindandbitcoin-clifor node operation, wallet encryption, and configuration management, which require significant system permissions and careful handling of credentials.
Audit Metadata