The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes multiple companion agents and automated commands that execute shell operations to manage the application lifecycle.
agents/cloudflare-deploy.md and commands/deploy.md automate the build and deployment process using npm run build, git commit, and npx wrangler deploy.
agents/worker-scaffold.md and commands/init.md initialize new projects by running npm install and creating directories/files.
agents/d1-migration.md manages database schemas using npx wrangler d1 execute.
[INDIRECT_PROMPT_INJECTION]: The cloudflare-debug agent includes a diagnostic step that tails real-time Worker logs, which represents an attack surface for indirect prompt injection.
Ingestion points: Untrusted data from web requests logged by the Worker is ingested into the agent context via npx wrangler tail in agents/cloudflare-debug.md.
Boundary markers: The skill does not define specific delimiters or instructions to ignore potential commands embedded in log data.
Capability inventory: The agent possesses Bash and Write capabilities, allowing it to execute commands or modify files based on its interpretation of the log content.
Sanitization: There is no evidence of filtering or sanitizing the log output before it is processed by the agent.

cloudflare-worker-base