Skip to content

jeremylongshore/tons-of-skills-marketplace / plugins/ai-ml/jeremy-gcp-starter-examples/skills/gcp-examples-expert

gcp-examples-expert

Verified

Generate production-ready Google Cloud code examples from official repositories including ADK samples, Genkit templates, Vertex AI notebooks, and Gemini patterns. Use when asked to "show ADK example" or "provide GCP starter kit". Trigger with relevant phrases based on skill purpose.

FileKindSize
SKILL.mdskill md6.1 KB
references/ARD.mdmarkdown5.2 KB
references/PRD.mdmarkdown4.2 KB
references/best-practices-applied.mdmarkdown1.2 KB
references/code-example-categories.mdmarkdown4.1 KB
references/errors.mdmarkdown4.4 KB
references/example-interactions.mdmarkdown1.5 KB
references/examples.mdmarkdown12.9 KB
references/workflow.mdmarkdown972 B
scripts/create-example.shexecutablescript2.0 KB

references/errors.md

View on GitHub
# Error Handling Reference

## Authentication & Authorization

| Error | Cause | Solution |
|-------|-------|----------|
| `UNAUTHENTICATED: Request had invalid authentication credentials` | ADC not configured or expired token | Run `gcloud auth application-default login`; verify `GOOGLE_APPLICATION_CREDENTIALS` env var points to a valid service account key |
| `PERMISSION_DENIED: caller does not have permission` | Service account missing required IAM roles | Grant `roles/aiplatform.user` for Vertex AI, `roles/run.developer` for Cloud Run; use `gcloud projects get-iam-policy` to audit bindings |
| `PERMISSION_DENIED: VPC Service Controls` | Request originates outside the VPC-SC perimeter | Add the caller's IP or access level to the perimeter; check `gcloud access-context-manager perimeters describe` |

## API & Quota

| Error | Cause | Solution |
|-------|-------|----------|
| `API [aiplatform.googleapis.com] not enabled` | Vertex AI API disabled on the project | Run `gcloud services enable aiplatform.googleapis.com` |
| `RESOURCE_EXHAUSTED: quota exceeded` | Rate limit or daily token quota hit | Request quota increase via Cloud Console; implement exponential backoff with jitter; consider batching requests |
| `429 Too Many Requests` | Gemini API rate-limited | Back off and retry; use `google-cloud-aiplatform` client library (auto-retries); switch lower-priority tasks to Gemini Flash |

## Model & Region

| Error | Cause | Solution |
|-------|-------|----------|
| `Model not found: gemini-2.5-pro` | Model not available in the specified region | Use `us-central1` or `europe-west4` where Gemini 2.5 models are generally available; check [regional availability docs](https://cloud.google.com/vertex-ai/generative-ai/docs/learn/locations) |
| `INVALID_ARGUMENT: unsupported model` | Using deprecated model name or wrong model ID | Verify model ID format: `gemini-2.5-pro`, `gemini-2.5-flash`; avoid legacy names like `gemini-1.0-pro` |
| `SAFETY_BLOCK` in response | Model safety filters triggered | Review prompt for policy violations; adjust safety settings if appropriate; add input sanitization layer |

## Dependency & Build

| Error | Cause | Solution |
|-------|-------|----------|
| `ModuleNotFoundError: No module named 'google.adk'` | ADK package not installed | Run `pip install google-adk`; verify virtual environment is activated |
| `Cannot find module '@genkit-ai/googleai'` | Genkit Google AI plugin not installed | Run `npm install @genkit-ai/googleai`; check `package.json` has correct version |
| `Version conflict` between AI SDK packages | Incompatible pinned versions | Pin all `@genkit-ai/*` packages to the same minor version; use lockfile (`package-lock.json` or `pnpm-lock.yaml`) |
| `ImportError: cannot import name 'Agent' from 'google.adk'` | Wrong import path for ADK Agent class | Use `from google.adk.agents import Agent` (not `from google.adk import Agent`) |

## Deployment

| Error | Cause | Solution |
|-------|-------|----------|
| `Cloud Run: Container failed to start` | Missing env vars or port misconfiguration | Ensure `PORT=8080` is set; verify `GOOGLE_GENAI_API_KEY` or credentials are passed via `--set-secrets` |
| `Cloud Run: Memory limit exceeded` | Container OOM during model inference | Increase memory with `--memory 1Gi` or higher; reduce batch sizes; use streaming responses |
| `Terraform: Error creating Agent` | Terraform provider version incompatible | Upgrade `hashicorp/google` provider to `~> 5.0`; run `terraform init -upgrade` |
| `adk deploy: timeout` | Large agent package or slow network | Increase timeout; verify `.gcloudignore` excludes `node_modules`, `venv`, and test data |

## Runtime

| Error | Cause | Solution |
|-------|-------|----------|
| `Timeout: request took longer than 300s` | Complex prompt or large context window | Reduce context size; use Gemini Flash for simple tasks; implement streaming for long generations |
| `INVALID_ARGUMENT: request payload too large` | Input exceeds model context limit | Chunk large documents; summarize before sending; use RAG to retrieve only relevant sections |
| `JSONDecodeError` when parsing structured output | Model returned non-JSON despite `response_mime_type: application/json` | Add explicit JSON schema in prompt; retry with lower temperature; validate output before parsing |

---
*[Tons of Skills](https://tonsofskills.com) by [Intent Solutions](https://intentsolutions.io) | [jeremylongshore.com](https://jeremylongshore.com)*
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 323kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall