The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides code templates that ingest untrusted data from HTTP requests and event triggers, creating a surface where malicious instructions in external data could influence the agent during development or testing cycles.
Ingestion points: req.body and req.query in SKILL.md (HTTP Function), cloudEvent.data in SKILL.md (Pub/Sub and Storage Functions).
Boundary markers: Absent. The provided templates do not demonstrate the use of delimiters or 'ignore embedded instructions' warnings for external data.
Capability inventory: The skill utilizes gcloud CLI tools and Node.js for networking and file operations across various patterns.
Sanitization: Absent. The templates show direct usage of input (e.g., req.query.name) or standard JSON.parse without validation logic.
[COMMAND_EXECUTION]: The skill includes instructions for deploying infrastructure and building containers using the Google Cloud CLI and Cloud Build.
The skill provides templates using gcloud run deploy, gcloud functions deploy, and gcr.io/cloud-builders/docker to manage serverless resources.
Deployment configurations in SKILL.md include the --allow-unauthenticated flag, which enables public access to the deployed services.

gcp-cloud-run