The Agent Skills Directory

[PROMPT_INJECTION]: The skill implements an architecture that ingests untrusted data from HTTP request bodies (req.json()) to trigger and parameterize workflows. While this is the intended design, it creates a surface for indirect prompt injection where malicious data could influence agent behavior if processed without strict validation. * Ingestion points: HTTP request handler in flows/index.ts and the start/ endpoint. * Boundary markers: None present in the code examples to delimit untrusted input. * Capability inventory: Access to Supabase database (supabase.from().upsert()) and Resonate workflow server. * Sanitization: The provided code patterns do not include explicit input sanitization or schema validation.- [SAFE]: The skill follows secure practices by using environment variables for sensitive credentials such as RESONATE_TOKEN and SUPABASE_SERVICE_ROLE_KEY via Deno.env.get(). It also uses well-known services and vendor-owned packages for its dependencies.

resonate-supabase-deployments-typescript