Loading
Also from Kynth Studios
Also from Kynth Studios
Also from Kynth Studios
1# LLM Package Guide23## Effect45- Prefer `HttpClient.HttpClient` / `HttpClientResponse.HttpClientResponse` over web `fetch` / `Response` at package boundaries.6- Use `Stream.Stream` for streaming data flow. Avoid ad hoc async generators or manual web reader loops unless an Effect `Stream` API cannot model the behavior.7- Use Effect Schema codecs for JSON encode/decode (`Schema.fromJsonString(...)`) instead of direct `JSON.parse` / `JSON.stringify` in implementation code.8- In `Effect.gen`, yield yieldable errors directly (`return yield* new MyError(...)`) instead of `Effect.fail(new MyError(...))`.9- Use `Effect.void` instead of `Effect.succeed(undefined)` when the successful value is intentionally void.1011## Conventions1213Per-type constructors live on the type, not as top-level re-exports. Use `Message.system(...)`, `Message.user(...)`, `Message.assistant(...)`, `Message.tool(...)`, `Model.make(...)`, `ToolDefinition.make(...)`, `ToolCallPart.make(...)`, `ToolResultPart.make(...)`, `ToolChoice.make(...)`, `ToolChoice.named(...)`, `SystemPart.make(...)`, and `GenerationOptions.make(...)` directly. The top-level `LLM` namespace is reserved for request-shaped call APIs: `LLM.request`, `LLM.generate`, `LLM.stream`, `LLM.updateRequest`, and `LLM.generateObject`. Two ways to construct the same thing is one too many.1415## Tests1617- Use `testEffect(...)` from `test/lib/effect.ts` for tests requiring Effect layers.18- Keep provider tests fixture-first. Live provider calls must stay behind `RECORD=true` and required API-key checks.1920## Architecture2122This package is an Effect Schema-first LLM core. The Schema classes in `src/schema/` are the canonical runtime data model. Convenience functions in `src/llm.ts` are thin constructors that return those same Schema class instances; they should improve callsites without creating a second model.2324Primary in-repo integration point:2526- `packages/opencode/src/session/llm.ts` is the session-owned orchestration layer that decides whether a request uses AI SDK or this package's native route runtime.27- `packages/opencode/src/session/llm/native-request.ts` is the lowering adapter from opencode's session/AI SDK-shaped data into this package's `LLMRequest` model.28- `packages/opencode/src/session/llm/native-runtime.ts` is the execution adapter that calls raw `LLMClient.stream(request)` and bridges one provider turn of opencode tool calls through this package's typed dispatcher.29- `packages/opencode/src/session/llm/ai-sdk.ts` keeps the default AI SDK path compatible by converting AI SDK stream parts into this package's shared `LLMEvent`s.3031Keep this package independent of session concerns. Session auth, permissions, plugins, telemetry headers, and runtime selection belong in `packages/opencode/src/session/llm.ts` and its local adapters.3233### Request Flow3435The intended callsite is:3637```ts38const request = LLM.request({39 model: OpenAI.configure({ apiKey }).responses("gpt-4o-mini"),40 system: "You are concise.",41 prompt: "Say hello.",42})4344const response = yield * LLMClient.generate(request)45```4647`LLM.request(...)` builds an `LLMRequest`. `LLMClient.generate(...)` reads the executable route carried by `request.model.route`, builds the provider-native body, asks the route's transport for a real `HttpClientRequest.HttpClientRequest`, sends it through `RequestExecutor.Service`, parses the provider stream into common `LLMEvent`s, and finally returns an `LLMResponse`.4849Use `LLMClient.stream(request)` when callers want incremental `LLMEvent`s. Use `LLMClient.generate(request)` when callers want those same events collected into an `LLMResponse`. Use `LLMClient.prepare<Body>(request)` to compile a request through the route pipeline without sending it — the optional `Body` type argument narrows `.body` to the route's native shape (e.g. `prepare<OpenAIChatBody>(...)` returns a `PreparedRequestOf<OpenAIChatBody>`). The runtime body is identical; the generic is a type-level assertion.5051Filter or narrow `LLMEvent` streams with `LLMEvent.is.*` (camelCase guards, e.g. `events.filter(LLMEvent.is.toolCall)`). The kebab-case `LLMEvent.guards["tool-call"]` form also works but prefer `is.*` in new code.5253### Routes5455A route is the registered, runnable composition of four orthogonal pieces:5657- **`Protocol`** (`src/route/protocol.ts`) — semantic API contract. Owns request body construction (`body.from`), the body schema (`body.schema`), the streaming-event schema (`stream.event`), and the event-to-`LLMEvent` state machine (`stream.step`). `Route.make(...)` validates and JSON-encodes the body from `body.schema` and decodes frames with `stream.event`. Examples: `OpenAIChat.protocol`, `OpenAIResponses.protocol`, `AnthropicMessages.protocol`, `Gemini.protocol`, `BedrockConverse.protocol`.58- **`Endpoint`** (`src/route/endpoint.ts`) — URL construction. The host, path, and route query live on the endpoint. `Endpoint.path("/chat/completions", { baseURL })` is the common case; pass a function for paths that embed the model id or a body field (e.g. `Endpoint.path(({ body }) => `/model/${body.modelId}/converse-stream`)`).59- **`Auth`** (`src/route/auth.ts`) — per-request transport authentication. Provider facades configure credentials onto the route before model selection, usually via `Auth.bearer(apiKey)` or `Auth.header(name, apiKey)`. Routes that need per-request signing (Bedrock SigV4, future Vertex IAM, Azure AAD) implement `Auth` as a function that signs the body and merges signed headers into the result.60- **`Framing`** (`src/route/framing.ts`) — bytes → frames. SSE (`Framing.sse`) is shared; Bedrock keeps its AWS event-stream framing as a typed `Framing<object>` value alongside its protocol.6162Compose them via `Route.make(...)`:6364```ts65export const route = Route.make({66 id: "openai-chat",67 provider: "openai",68 protocol: OpenAIChat.protocol,69 endpoint: Endpoint.path("/chat/completions", {70 baseURL: "https://api.openai.com/v1",71 }),72 auth: Auth.bearer(),73 framing: Framing.sse,74})75```7677Route defaults are request-shaping defaults such as `headers`, `limits`, `generation`, `providerOptions`, and `http`. Endpoint host/query belongs on the route endpoint. Selected `Model` values carry only model id, provider id, and the configured route value. Model capability/catalog metadata lives outside this package; protocol support is enforced by request lowering and typed `LLMError`s.7879The four-axis decomposition is the reason DeepSeek, TogetherAI, Cerebras, Baseten, Fireworks, and DeepInfra all reuse `OpenAIChat.protocol` verbatim — each provider deployment is a 5-15 line `Route.make(...)` call instead of a 300-400 line route clone. Bug fixes in one protocol propagate to every consumer of that protocol in a single commit.8081When a provider ships a non-HTTP transport (OpenAI's WebSocket Responses backend, hypothetical bidirectional streaming APIs), the seam is `Transport` — `WebSocketTransport.jsonTransport.with(...)` constructs an IO template whose `prepare` receives the route endpoint/auth at compile time, builds a WebSocket URL and message, and whose `frames` yields decoded text from the socket. Same protocol and endpoint source, different transport.8283### URL Construction8485`Endpoint` owns `{ baseURL, path, query }`. Each protocol route includes a canonical endpoint when the provider has one (e.g. `https://api.openai.com/v1`); provider helpers override endpoint fields by configuring the route before selecting a model. Routes that have no canonical URL (OpenAI-compatible Chat, GitHub Copilot) require configuration before execution.8687For providers where the URL is derived from typed inputs (Azure resource name, Bedrock region), the provider helper configures the route endpoint before calling `.model(...)`. Use `AtLeastOne<T>` from `route/auth-options.ts` for inputs that accept either of two derivation paths (Azure: `resourceName` or `baseURL`).8889### Provider Facades9091Provider-facing APIs are configured facades over route values. Endpoint/auth/resource/API-version setup happens before model selection, and model selectors accept only a model or deployment id:9293```ts94const openai = OpenAI.configure({ apiKey, baseURL })95const model = openai.responses("gpt-4o-mini")9697const azure = Azure.configure({ resourceName, apiKey, apiVersion: "v1" })98const deployment = azure.responses("my-deployment")99100const gateway = CloudflareAIGateway.configure({ accountId, gatewayId, gatewayApiKey, apiKey })101const proxied = gateway.model("openai/gpt-4o-mini")102```103104Keep provider facades small and explicit:105106- Use branded `ProviderID.make(...)` and `ModelID.make(...)` where ids are constructed directly.107- Use `model` for the default API path and named methods for provider-native alternatives such as OpenAI `responses`, `responsesWebSocket`, and `chat`.108- Put provider-specific setup on `.configure(...)`; do not add `model(id, overrides)` as a duplicate construction path.109- Export lower-level `routes` arrays separately only when advanced internal wiring needs them.110- Prefer `apiKey` as provider-specific sugar and `auth` as the explicit override; keep them mutually exclusive in provider option types with `ProviderAuthOption`.111- Resolve `apiKey` → `Auth` with `AuthOptions.bearer(options, "<PROVIDER>_API_KEY")` (it honors an explicit `auth` override and falls back to `Auth.config(envVar)` so missing keys surface a typed `Authentication` error rather than a runtime crash).112- Use separate top-level facades for products with different required setup, such as `CloudflareAIGateway` and `CloudflareWorkersAI`.113114`Provider.make(...)` remains available for simple static provider definitions, but new built-in providers should prefer plain configured facades unless a helper removes real duplication without adding runtime behavior.115116### Folder layout117118```119packages/llm/src/120 schema/ canonical Schema model, split by concern121 ids.ts branded IDs, literal types, ProviderMetadata122 options.ts Generation/Provider/Http options, Limits, Model, cache policy123 messages.ts content parts, Message, ToolDefinition, LLMRequest124 events.ts Usage, individual events, LLMEvent, PreparedRequest, LLMResponse125 errors.ts error reasons, LLMError, ToolFailure126 index.ts barrel127 llm.ts request constructors and convenience helpers128 route/129 index.ts @opencode-ai/llm/route advanced barrel130 client.ts Route.make + LLMClient.prepare/stream/generate131 executor.ts RequestExecutor service + transport error mapping132 protocol.ts Protocol type + Protocol.make133 endpoint.ts Endpoint type + Endpoint.path134 auth.ts Auth type + Auth.bearer / Auth.apiKeyHeader / Auth.passthrough135 auth-options.ts ProviderAuthOption shape, AuthOptions.bearer, AtLeastOne helper136 framing.ts Framing type + Framing.sse137 transport/ transport implementations138 index.ts Transport type + HttpTransport / WebSocketTransport namespaces139 http.ts HttpTransport.httpJson — POST + framing140 websocket.ts WebSocketTransport.json + WebSocketExecutor service141 protocols/142 shared.ts ProviderShared toolkit used inside protocol impls143 openai-chat.ts protocol + route (compose OpenAIChat.protocol)144 openai-responses.ts145 anthropic-messages.ts146 gemini.ts147 bedrock-converse.ts148 bedrock-event-stream.ts framing for AWS event-stream binary frames149 openai-compatible-chat.ts route that reuses OpenAIChat.protocol, no canonical URL150 utils/ per-protocol helpers (auth, cache, media, tool-stream, ...)151 providers/152 openai-compatible.ts generic compatible helper + family model helpers153 openai-compatible-profile.ts family defaults (deepseek, togetherai, ...)154 azure.ts / amazon-bedrock.ts / cloudflare.ts / github-copilot.ts / google.ts / xai.ts / openai.ts / anthropic.ts / openrouter.ts155 tool.ts typed tool() helper156 tool-runtime.ts narrow one-call typed tool dispatcher157```158159The dependency arrow points down: `providers/*.ts` files import protocol routes and auth-option utilities; protocol modules import `endpoint`, `auth`, `framing`, and transport pieces. Protocols do not import provider facades. Lower-level modules know nothing about provider catalog metadata.160161### Shared protocol helpers162163`ProviderShared` exports a small toolkit used inside protocol implementations to keep them focused on provider-native shapes:164165- `joinText(parts)` — joins an array of `TextPart` (or anything with a `.text`) with newlines. Use this anywhere a protocol flattens text content into a single string for a provider field.166- `parseToolInput(route, name, raw)` — Schema-decodes a tool-call argument string with the canonical "Invalid JSON input for `<route>` tool call `<name>`" error message. Treats empty input as `{}`.167- `parseJson(route, raw, message)` — generic JSON-via-Schema decode for non-tool bodies.168- `eventError(route, message, ...)` — typed `InvalidProviderOutput` constructor for stream-time decode failures.169- `validateWith(decoder)` — maps Schema decode errors to `InvalidRequest`. `Route.make(...)` uses this for body validation; lower-level routes can reuse it.170- `matchToolChoice(provider, choice, branches)` — branches over `LLMRequest["toolChoice"]` for provider-specific lowering.171172If you find yourself copying a 3-to-5-line snippet between two protocols, lift it into `ProviderShared` next to these helpers rather than duplicating.173174### Chronological System Updates175176`LLMRequest.system` is the initial privileged prompt that applies ahead of the conversation. `Message.system(...)` is a separate, provider-neutral chronological operator update inside `LLMRequest.messages`; it applies only from its position in history onward and accepts text content only.177178Native chronological system messages are route/model-specific. Anthropic Messages lowers them natively for Claude Opus 4.8 (`claude-opus-4-8`). Other routes and models intentionally lower the update in place into ordinary user-compatible text using this stable escaped representation:179180```text181<system-update>182...183</system-update>184```185186The wrapped-user fallback preserves ordering while visibly lowering authority. Never silently pass a raw chronological `role: "system"` through a route that might reject it. Do not insert raw retrieved documents, tool output, or web content into privileged chronological system updates; keep untrusted content in ordinary user/tool channels.187188### Tools189190Tool loops are represented in common messages and events:191192```ts193const call = ToolCallPart.make({ id: "call_1", name: "lookup", input: { query: "weather" } })194const result = Message.tool({ id: "call_1", name: "lookup", result: { forecast: "sunny" } })195196const followUp = LLM.request({197 model,198 messages: [Message.user("Weather?"), Message.assistant([call]), result],199})200```201202Routes lower these into provider-native assistant tool-call messages and tool-result messages. Streaming providers should emit `tool-input-delta` events while arguments arrive, then a final `tool-call` event with parsed input.203204### Tool dispatch205206`LLM.stream(request)` and `LLM.generate(request)` each run exactly one provider turn. Add tool schemas to `request.tools` with `Tool.toDefinitions(tools)`. When a caller wants the package's typed one-call execution behavior, pass each canonical local `tool-call` event to `ToolRuntime.dispatch(tools, call)`.207208```ts209const get_weather = tool({210 description: "Get current weather for a city",211 parameters: Schema.Struct({ city: Schema.String }),212 success: Schema.Struct({ temperature: Schema.Number, condition: Schema.String }),213 execute: ({ city }) =>214 Effect.gen(function* () {215 // city: string — typed from parameters Schema216 const data = yield* WeatherApi.fetch(city)217 return { temperature: data.temp, condition: data.cond }218 // return type checked against success Schema219 }),220})221222const tools = { get_weather, get_time, ... }223const events = yield* LLM.stream(224 LLM.updateRequest(request, { tools: Tool.toDefinitions(tools) }),225).pipe(Stream.runCollect)226227const call = Array.from(events).find(LLMEvent.is.toolCall)228if (call && !call.providerExecuted) {229 const dispatched = yield* ToolRuntime.dispatch(tools, call)230 // Persist call + dispatched.result, then construct the next request explicitly.231}232```233234The dispatcher:235236- On `tool-call`: looks up the named tool, decodes input against `parameters` Schema, dispatches to the typed `execute`, encodes the result against `success` Schema, and returns canonical `tool-result` events.237- Does not stream providers, construct Session events, schedule fibers, append history, count steps, or continue model rounds.238- Leaves persistence and continuation to the enclosing product flow.239240Handler dependencies (services, permissions, plugin hooks, abort handling) are closed over by the consumer at tool-construction time. Build the tools record inside an `Effect.gen` once and reuse it across many dispatches.241242Errors must be expressed as `ToolFailure`. The runtime catches it and emits a `tool-error` event, then a `tool-result` of `type: "error"`, so the model can self-correct on the next step. Anything that is not a `ToolFailure` is treated as a defect and fails the stream. Three recoverable error paths produce `tool-error` events:243244- The model called an unknown tool name.245- Input failed the `parameters` Schema.246- The handler returned a `ToolFailure`.247248Provider-defined / hosted tools (Anthropic `web_search` / `code_execution` / `web_fetch`, OpenAI Responses `web_search_call` / `file_search_call` / `code_interpreter_call` / `mcp_call` / `local_shell_call` / `image_generation_call` / `computer_use_call`) pass through the runtime untouched:249250- Routes surface the model's call as a `tool-call` event with `providerExecuted: true`, and the provider's result as a matching `tool-result` event with `providerExecuted: true`.251- Callers detect `providerExecuted` on `tool-call` and **skip local dispatch** — no handler is invoked and no `tool-error` is raised for "unknown tool". The provider already executed it.252- Callers that continue should retain both events in explicit history when the protocol requires it. Anthropic encodes them back as `server_tool_use` + `web_search_tool_result` (or `code_execution_tool_result` / `web_fetch_tool_result`) blocks; OpenAI Responses callers typically use `previous_response_id` instead of resending hosted-tool items.253254Add provider-defined tools to `request.tools` (no runtime entry needed). The matching route must know how to lower the tool definition into the provider-native shape; right now Anthropic accepts `web_search` / `code_execution` / `web_fetch` and OpenAI Responses accepts the hosted tool names listed above.255256## Protocol File Style257258Protocol files should look self-similar. Provider quirks belong behind named helpers so a new route can be reviewed by comparing the same sections across files.259260### Section order261262Use this order for every protocol module:2632641. Public model input2652. Request body schema2663. Streaming event schema2674. Parser state2685. Request body construction (`fromRequest`)2696. Stream parsing (`step` and per-event handlers)2707. Protocol and route2718. Protocol route export272273### Rules274275- Keep protocol files focused on the protocol. Move provider-specific projection, signing, media normalization, or other bulky transformations into `src/protocols/utils/*`.276- Use `Effect.fn("Provider.fromRequest")` for request body construction entrypoints. Use `Effect.fn(...)` for event handlers that yield effects; keep purely synchronous handlers as plain functions returning a `StepResult` that the dispatcher lifts via `Effect.succeed(...)`.277- Parser state owns terminal information. The state machine records finish reason, usage, and pending tool calls; emit one terminal `finish` event (or `provider-error`) for each completed response. If a provider splits reason and usage across events, merge them in parser state before flushing.278- Emit exactly one terminal `finish` event for a completed response, normally after a matching `step-finish`. Use `stream.terminal` to stop reading when the provider has a completion sentinel; use `stream.onHalt` when the final event must be flushed after the framed stream ends.279- Use shared helpers for repeated protocol policy such as text joining, usage totals, JSON parsing, and tool-call accumulation. `ToolStream` (`protocols/utils/tool-stream.ts`) accumulates streamed tool-call arguments uniformly.280- Make intentional provider differences explicit in helper names or comments. If two protocol files differ visually, the reason should be obvious from the names.281- Prefer dispatched per-event handlers (`onMessageStart`, `onContentBlockDelta`, ...) called from a small top-level `step` switch over a long if-chain. The dispatcher keeps the event surface visible at a glance.282- Keep tests in the same conceptual order as the protocol: basic prepare, tools prepare, unsupported lowering, text/usage parsing, tool streaming, finish reasons, provider errors.283284### Review checklist285286- Can the file be skimmed side-by-side with `openai-chat.ts` without hunting for equivalent sections?287- Are provider quirks named, isolated, and covered by focused tests?288- Does request body construction validate unsupported common content at the protocol boundary?289- Does stream parsing emit stable common events without leaking provider event order to callers?290- Does `toolChoice: "none"` behavior read as intentional?291292## Recording Tests293294Recorded tests use one cassette file per scenario. A cassette holds an ordered array of `{ request, response }` interactions, so multi-step flows (tool loops, retries, polling) record into a single file. Use `recordedTests({ prefix, requires })` and let the helper derive cassette names from test names:295296```ts297const recorded = recordedTests({ prefix: "openai-chat", requires: ["OPENAI_API_KEY"] })298299recorded.effect("streams text", () =>300 Effect.gen(function* () {301 // test body302 }),303)304```305306Replay is the default. `RECORD=true` records fresh cassettes and requires the listed env vars. Cassettes are written as pretty-printed JSON so multi-interaction diffs stay reviewable.307308Pass `provider`, `protocol`, and optional `tags` to `recordedTests(...)` / `recorded.effect.with(...)` so cassettes carry searchable metadata. Use recorded-test filters to replay or record a narrow subset without rewriting a whole file:309310- `RECORDED_PROVIDER=openai` matches tests tagged with `provider:openai`; comma-separated values are allowed.311- `RECORDED_PREFIX=openai-chat` matches cassette groups by `recordedTests({ prefix })`; comma-separated values are allowed.312- `RECORDED_TAGS=tool` requires all listed tags to be present, e.g. `RECORDED_TAGS=provider:togetherai,tool`.313- `RECORDED_TEST="streams text"` matches by test name, kebab-case test id, or cassette path.314315Filters apply in replay and record mode. Combine them with `RECORD=true` when refreshing only one provider or scenario.316317**Binary response bodies.** Most providers stream text (SSE, JSON). The recorder treats known textual media types (`text/*`, JSON/XML structured types, JavaScript, forms, YAML, and SVG) as text and stores every other response as base64 with `bodyEncoding: "base64"`. This preserves binary formats such as AWS event-stream frames without a lossy UTF-8 round trip.318319**Matching strategy.** Replay walks the cassette in record order via an internal cursor: the Nth runtime request is served by the Nth recorded interaction, and each one is validated by comparing method, URL, allow-listed headers, and the canonical JSON body. This handles tool loops (each round's request differs as history grows) and retry/polling scenarios (successive byte-identical requests with different responses) uniformly. If a test reorders its requests, re-record the cassette. `scriptedResponses` (in `test/lib/http.ts`) is the deterministic counterpart for tests that don't need a live provider; it scripts response bodies in order without reading from disk.320321Do not blanket re-record an entire test file when adding one cassette. `RECORD=true` rewrites every recorded case that runs, and provider streams contain volatile IDs, timestamps, fingerprints, and obfuscation fields. Prefer deleting the one cassette you intend to refresh, or run a focused test pattern that only registers the scenario you want to record. Keep stable existing cassettes unchanged unless their request shape or expected behavior changed.322
One repository carrying more than one format is the comparison this product exists for: does anyone actually write different content in each file, or is one a copy of the other?
| Repository | Format | Stack | Covers | Score | Changed |
|---|---|---|---|---|---|
| anomalyco/opencodepackages/opencode/src/server/routes/instance/httpapi/AGENTS.md · 198k | AGENTS.md | styleapi | 40/100 | 14 days ago | |
| anomalyco/opencodeAGENTS.md · 198k | AGENTS.md | testlint-formatstyletypes+3 | 80/100 | 14 days ago | |
| anomalyco/opencodepackages/app/AGENTS.md · 198k | AGENTS.md | style | 60/100 | 8 days ago | |
| anomalyco/opencodepackages/app/e2e/AGENTS.md · 198k | AGENTS.md | teststyledo-not | 54/100 | 9 days ago | |
| anomalyco/opencodepackages/app/e2e/performance/AGENTS.md · 198k | AGENTS.md | no sections | 16/100 | 14 days ago | |
| anomalyco/opencodepackages/codemode/AGENTS.md · 198k | AGENTS.md | api | 43/100 | 14 days ago | |
| anomalyco/opencodepackages/core/src/tool/AGENTS.md · 198k | AGENTS.md | stylesecurity | 58/100 | 14 days ago | |
| anomalyco/opencodepackages/desktop/AGENTS.md · 198k | AGENTS.md | style | 34/100 | 8 days ago | |
| anomalyco/opencodepackages/effect-drizzle-sqlite/AGENTS.md · 198k | AGENTS.md | database | 38/100 | 14 days ago | |
| anomalyco/opencodepackages/opencode/test/AGENTS.md · 198k | AGENTS.md | teststylearchtesting-strategy+1 | 81/100 | 14 days ago | |
| anomalyco/opencodepackages/opencode/test/server/AGENTS.md · 198k | AGENTS.md | teststyleapi | 38/100 | 14 days ago | |
| anomalyco/opencodepackages/schema/AGENTS.md · 198k | AGENTS.md | styletypesdatabaseapi+1 | 65/100 | 14 days ago | |
| anomalyco/opencodepackages/session-ui/AGENTS.md · 198k | AGENTS.md | no sections | 30/100 | 11 days ago | |
| anomalyco/opencodepackages/ui/AGENTS.md · 198k | AGENTS.md | style | 34/100 | 8 days ago | |
| anomalyco/opencodepackages/opencode/AGENTS.md · 198k | AGENTS.md | styledatabaseapido-not | 81/100 | 14 days ago | |
| anomalyco/opencodepackages/opencode/src/session/llm/AGENTS.md · 198k | AGENTS.md | arch | 61/100 | 14 days ago |
Same format, overlapping stack, ranked by quality.
| Repository | Format | Stack | Covers | Score | Changed |
|---|---|---|---|---|---|
| TryGhost/Ghoste2e/AGENTS.md · 55k | AGENTS.md | setupteststylearch+2 | 100/100 | today | |
| mui/material-uiAGENTS.md · 99k | AGENTS.md | setupbuildtestlint-format+9 | 100/100 | 14 days ago | |
| code-yeongyu/oh-my-openagentpackages/web/AGENTS.md · 68k | AGENTS.md | setupbuildtestlint-format+6 | 100/100 | 13 days ago | |
| aaif-goose/gooseAGENTS.md · 53k | AGENTS.md | setupbuildtestlint-format+7 | 100/100 | 8 days ago | |
| deepseek-ai/deepseek-harnessnative/landlock-run/AGENTS.md · 104k | AGENTS.md | setupteststylearch+3 | 100/100 | today | |
| duckduckgo/content-scope-scriptsspecial-pages/AGENTS.md · 70 | AGENTS.md | buildteststylearch+3 | 100/100 | 14 days ago | |
| n8n-io/n8npackages/@n8n/agents/AGENTS.md · 201k | AGENTS.md | buildteststylearch+3 | 100/100 | 14 days ago | |
| elastic/elasticsearchx-pack/plugin/core/src/main/java/org/elasticsearch/xpack/core/ml/AGENTS.md · 78k | AGENTS.md | buildtestlint-formatstyle+2 | 100/100 | 14 days ago |
A badge carrying the measured quality of the strongest agent config file in this repository, out of 100. It reads from this index every time somebody loads your page, so it changes when the measurement changes and there is nothing to keep up to date. Free, no account, and the value is not something you or we can set by hand.
[](https://rulestack.kynth.studio/configs/anomalyco-opencode-packages-llm-agents)Would rather not hotlink us? Every badge is also served in shields.io’s endpoint schema, so shields renders the image and your readers never talk to our domain:
Published by Toolproof, the masthead over this index and eight others. The method behind the number is at toolproof.kynth.studio/methodology, and the whole thing is readable as JSON with no key at /api.