The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads and executes the @builtbyecho/agent-wormhole package from npm via npx. This represents an external dependency from the provider (BuiltByEcho). The skill includes a mandatory verification step where the agent must ask for user consent before the first execution in a session.
[COMMAND_EXECUTION]: Shell commands are used to invoke the agent-wormhole CLI for sending, receiving, and inspecting encrypted payloads, as well as the bankr CLI for verifying x402 payment status at the x402.bankr.bot endpoint.
[DATA_EXFILTRATION]: Encrypted mission briefs, artifacts, and secrets are transmitted to storage.builtbyecho.xyz. This is the core functionality of the skill for remote agent handoffs, and data is protected by local AES-256-GCM encryption prior to transmission.
[PROMPT_INJECTION]: The skill processes untrusted data received through the handoff mechanism, creating an attack surface for indirect prompt injection.
Ingestion points: agent-wormhole receive writes external content to the ./received directory.
Boundary markers: None specified in the instructions.
Capability inventory: The skill can execute shell commands (npx, bankr) and perform file system operations.
Sanitization: No explicit sanitization of received content is mentioned before storage.

agent-wormhole