The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (npm run crawl:fixtures, npm run sync:fixtures) using interpolated arguments such as <team-slug> and <league-name>. If these inputs are not strictly sanitized by the execution environment, they could serve as a vector for command injection.
[EXTERNAL_DOWNLOADS]: The skill automates the retrieval of data from fv.dribl.com using Playwright. Although the source is a legitimate sports management platform, the data is externally controlled and untrusted.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) due to its data ingestion model.
Ingestion points: External data is scraped from fv.dribl.com and saved to the local file system in data/external/ and data/matches/.
Boundary markers: The skill instructions do not define delimiters or specific warnings to the agent regarding the separation of ingested data from operational instructions.
Capability inventory: The skill possesses capabilities for network access (scraping), file system modifications (writing JSON outputs), and execution of local repository scripts.
Sanitization: The skill employs Zod validation schemas (src/types/matches.ts) to enforce data structure, which provides a layer of defense against malformed data but does not inherently filter malicious natural language instructions embedded within the scraped content.

wsc-fixture-pipeline