The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the vendor's CLI tool to interact with Polymarket.
Evidence: Use of npx @useclawlink/cli login, connect, and run commands in SKILL.md to manage authentication and execute trading actions.
[EXTERNAL_DOWNLOADS]: The skill downloads and executes the CLI package from the NPM registry at runtime using npx.
Evidence: Reference to @useclawlink/cli in SKILL.md for installation and execution.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from Polymarket (such as market descriptions, event names, and order books) which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Market search, market details, and event discovery actions defined in SKILL.md.
Boundary markers: The skill includes a specific instruction: "Reads first; confirm with the user before any write."
Capability inventory: The agent has the capability to create and cancel orders (polymarket_create_order, polymarket_cancel_order) via the CLI.
Sanitization: No explicit technical sanitization or schema validation of the Polymarket data is described in the skill instructions.

polymarket