Maintain an application's supported language models by checking provider availability, adding models, and keeping documentation, live tests, and model-specific request settings aligned. It also checks provider model identifiers and candidates for the recommended list.
中文简介正在准备中。当前显示英语简介。
Maintain the supported LLM model list: add a new model, or run routine maintenance to verify availability and discover new models worth adding. Use when the user asks to add/support a model, update the model list, or check model availability.
安装前可以查看交给 AI 代理的指令内容。
The supported model list lives in three places that must stay in sync:
| File | What it holds |
|---|---|
packages/website/src/pages/docs/features/models/page.tsx | MODEL_GROUPS (public docs, source of truth for display names) and BASELINE (recommended models) |
packages/llms/src/models.live.test.ts | Mirrored MODEL_GROUPS, plus OPENROUTER_ID_OVERRIDES / ALIYUN_ID_OVERRIDES for provider-specific model ids |
packages/llms/src/utils.ts | modelPatch — per-model-family request parameter fixes (thinking/reasoning flags, tool_choice quirks, etc.) |
git status. If the working tree has uncommitted changes that look like unrelated in-progress work, do NOT edit files — report the situation and ask the user for permission first.npm run test:live -w @page-agent/llms. API keys come from the repo-root .env (TESTING_OPENROUTER_KEY, TESTING_ALIYUN_KEY, TESTING_DEEPSEEK_KEY); tests skip silently when a key is missing, so check which providers actually ran. Record which models pass/fail before making changes, so new failures are attributable.https://openrouter.ai/api/v1/models and find the exact id (<vendor-slug>/<model-id>, watch for -preview, dated snapshots, dots vs hyphens).tool_choice, parallel_tool_calls, or parameter-schema quirks? Does the vendor's OpenAI-compatible endpoint differ from OpenRouter's behavior?modelPatch in packages/llms/src/utils.ts if the model needs new parameter handling. Model names are matched after normalizeModelName (lowercased, /-prefix, . and _ stripped) — check whether an existing family branch already covers it.MODEL_GROUPS lists (website page and live test), newest first within its brand group. Add id overrides if the OpenRouter/Aliyun id differs from the display name.npm run test:live -w @page-agent/llms
A failure means the request/response shape is wrong for that model — fix modelPatch or the id override, don't shrug it off. If the model fails consistently on a provider, exclude it from that provider (see the deepseek-3.2 precedent in the test file) and document why in a comment. 5. Decide on BASELINE only if the model is a fast, cheap, strong-tool-call option; otherwise leave it out. 6. Run npm run typecheck and npm test.
OPENROUTER_ID_OVERRIDES against the current https://openrouter.ai/api/v1/models output; ids change when vendors promote previews to stable.MODEL_GROUPS, plus notable newcomers). A candidate must support tool calls via an OpenAI-compatible API.After your own tests pass, always remind the user to verify manually:
.env — give the exact model id per channel, including the OpenRouter id, e.g.:LLM_BASE_URL="https://openrouter.ai/api/v1"
LLM_API_KEY="..."
LLM_MODEL_NAME="<exact-openrouter-id>"
npm run dev:demo (serves on port 5174) and exercise the agent against a real page.The live test only proves a single forced tool call round-trips; it is not an agent-quality eval. The user's manual run is the real acceptance test.
还没有评价。欢迎分享你的使用感受。
无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。
还没有评论。来写第一条吧。
Clean up local Git branches and remote connections left after pull request reviews. The workflow checks unfinished work, preserves main and the current branch, and requests confirmation before force-deleting branches that have not been merged.
中文简介正在准备中。当前显示英语简介。
Evaluate a proposed software change before implementation. It examines project files and official sources, compares options and trade-offs, and builds a concrete plan covering affected systems, effort, risks, and unresolved questions for the user's approval.
中文简介正在准备中。当前显示英语简介。
Submit existing local changes as a pull request by reviewing the diff, running project checks, preparing a branch and commit, and pushing to GitHub. It follows contribution rules and preserves template declarations that require human verification.
中文简介正在准备中。当前显示英语简介。
Update a project's changelog using GitHub releases, git history, and code differences. Select changes that matter to users, match the existing format, and write concise entries for the requested release.
中文简介正在准备中。当前显示英语简介。