The Agent Skills Directory

[PROMPT_INJECTION]: The skill includes an 'Execution Rule' in SKILL.md that directs the agent to 'Run the user's command directly without pre-checking; fix on failure.' This instruction attempts to override the agent's standard caution and verification protocols.
[PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by fetching raw JSON data from an external API and providing it to the agent without sanitization or boundary markers.
Ingestion points: Data is fetched from the external POETRY_API_URL (defaulting to https://poetry.palemoky.com/api) in scripts/main.ts.
Boundary markers: None; the response is printed verbatim as raw JSON to the agent's context.
Capability inventory: The script uses execSync for system configuration and performs network requests.
Sanitization: No validation or escaping is performed on the data returned by the API before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill performs network requests to an external API service. The default endpoint is https://poetry.palemoky.com/api, which is the official demo instance for the associated open-source project.
[COMMAND_EXECUTION]: The scripts/main.ts script executes execSync("chcp 65001") on Windows platforms to configure UTF-8 encoding.
[COMMAND_EXECUTION]: The SKILL.md file contains instructions for the agent to execute a Docker command (docker run -d -p 1279:1279 palemoky/chinese-poetry-api:latest) to start a local instance of the API server if a reachable instance is not found.

chinese-poetry-api