Example User Prompts
Use these to test ModelPort triggering and behavior.
Basic migrations
- “Migrate all model usage in
src/agents/ from our old model config to the new release, keep tool behavior stable, and give me a rollback plan.”
- “Upgrade this codebase from provider A to provider B and adjust prompts so output format remains identical.”
- “Replace deprecated model parameters in our API callers and verify tests still pass.”
- “Update agent orchestration prompts and tool descriptions for the new model behavior, but do not change business logic.”
- “Do a phased migration in
services/search with side-by-side support for old and new models.”
- “Port all prompt templates in
prompts/ and flag any risky behavior changes.”
- “Migrate this repo to the latest model family and separate required fixes vs optional tuning.”
- “Upgrade model references across config, tests, and docs without breaking runtime routing.”
Current-model migrations
- “Migrate this repo to
claude-opus-5-5. Apply required fixes first, then list optional tuning separately.”
- “Move the classification and routing calls in
src/ to claude-haiku-5-5 and keep the output format identical.”
- “Migrate our OpenAI calls from
gpt-5.5 to gpt-6.1-sol. Move tool calling to the Responses API where needed.”
- “Upgrade our Gemini calls to
gemini-3.8-flash and replace thinking_budget with thinking_level.”
- “We changed the model ID and now get 400s. Find the cause and fix it.”
- “Our agent UI goes quiet between tool calls since the upgrade. Find out why and fix it.”
Advanced migrations
- “Migrate from OpenAI
gpt-5.5 to Anthropic claude-sonnet-5-5 across all agents, normalize output contracts, and add provider-agnostic wrapper layers.”
- “Agent B already started migrating
services/chat — pick up where it left off, verify what’s done, and finish the remaining work with validation.”
- “Migrate our structured output pipeline from JSON mode to native structured outputs on the new model, update Zod schemas, and regression-test all tool call paths.”