The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes the @openserv-labs/client package from the official NPM registry and references repositories under the openserv-labs organization.
[COMMAND_EXECUTION]: Provides instructions for using vendor-specific CLI tools such as npx @openserv-labs/client deploy for agent deployment and npx skills for maintenance.
[DATA_EXFILTRATION]: Facilitates communication with official platform infrastructure at api.openserv.ai and platform.openserv.ai for management and trigger operations.
[INDIRECT_PROMPT_INJECTION]: The skill defines an architecture where agents ingest untrusted data from external sources (webhooks, x402 payments).
Ingestion points: Webhook and x402 trigger input fields defined in SKILL.md and examples/agent.ts.
Boundary markers: Absent; no delimiters or instructions are provided to the LLM to ignore potentially malicious content within the ingested payload.
Capability inventory: The example agent in examples/agent.ts includes a processRequest capability which operates on this untrusted input.
Sanitization: Absent; the documentation does not describe or implement input validation or sanitization for data received via triggers.

openserv-client