The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructs the agent to read local files that typically contain sensitive secrets and credentials, specifically .env, .dev.vars, wrangler.toml, and GitHub Actions workflow files in .github/workflows/.
[CREDENTIALS_UNSAFE]: The skill directs the agent to execute wrangler secret list to retrieve and display configured secrets from the Cloudflare worker environment.
[COMMAND_EXECUTION]: The skill relies on the execution of multiple system-level CLI tools and package managers, including wrangler, gh, npx, pnpm, detox, curl, openssl, and dig, to collect system information and perform diagnostic tests.
[COMMAND_EXECUTION]: The skill suggests running build and test commands like npx expo prebuild --clean and npx stryker run, which execute code from the local project and its dependencies.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from logs and network requests while having significant system capabilities. (Ingestion points: wrangler tail, system logs (iOS/Android), and browser console/network logs; Boundary markers: Absent; Capability inventory: Access to sensitive files and various system commands; Sanitization: Absent).

debug-runbook