Repository navigation
Conversation
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 12, 2026 00:00
f34d624 to
7409202
Compare
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 12, 2026 00:09
7409202 to
af2c4ab
Compare
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 12, 2026 00:49
af2c4ab to
dff7cd3
Compare
allozaur
changed the base branch from
allozaur/refactor/ui-folders
to
allozaur/style/glass-surfaces
November 12, 2026 01:28
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 12, 2026 01:29
dff7cd3 to
7fd2f56
Compare
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
3 times, most recently
from
November 12, 2026 14:29
c1882f0 to
be6fe36
Compare
allozaur
changed the base branch from
allozaur/style/glass-surfaces
to
allozaur/feat/prompt-library
November 12, 2026 20:54
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 12, 2026 20:55
be6fe36 to
a48ffd4
Compare
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 12, 2026 22:59
a48ffd4 to
8f887f0
Compare
allozaur
changed the base branch from
allozaur/feat/prompt-library
to
allozaur/style/glass-surfaces
November 12, 2026 22:59
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 16, 2026 13:44
053f0c8 to
3eb58bb
Compare
allozaur
changed the base branch from
allozaur/style/glass-surfaces
to
allozaur/docs/llama-ui-storybook
November 16, 2026 13:45
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 16, 2026 13:52
3eb58bb to
23f5ffa
Compare
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 16, 2026 16:12
23f5ffa to
64799b7
Compare
allozaur
changed the base branch from
allozaur/docs/llama-ui-storybook
to
allozaur/feat/llama-chat-npx
November 16, 2026 16:13
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 16, 2026 16:50
64799b7 to
b285cda
Compare
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 16, 2026 17:36
b285cda to
f98f578
Compare
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 16, 2026 18:58
f98f578 to
432435d
Compare
allozaur
force-pushed
the
allozaur/refactor/llama-harness-package
branch
from
November 17, 2026 05:21
432435d to
f69ce45
Compare
This was referenced Nov 17, 2026
…/lib/harness Llama Harness is the agentic loop as headless TypeScript with no UI, no DOM and no dependencies. It streams a turn, runs the tools the model called, gives it the results, and repeats until the model answers without a tool call. An app gives it an inference client and tools, and reads the run as a stream of events: run(messages, tools, signal). It ships a client for any OpenAI-compatible endpoint, so it also runs on its own in Node. The UI workflow type-checks, tests and packs it. Assisted-by: Claude (Fable 5.2)
The loop leaves the agentic store, which now gives the harness what the app owns: ChatService as its inference client, the enabled tools behind the permission gate, and the callbacks that store each turn. Its API and the flow's callbacks are unchanged. A unit test runs the flow end to end with the server and the tools mocked. Chat compiles the harness from source through an alias, so the builds that copy or isolate ui/apps/chat now bring ui/lib/harness along: CMake and Nix through LLAMA_HARNESS_DIR, the Dockerfiles with a COPY. Assisted-by: Claude (Fable 5.2)
The layout card no longer calls the move planned, the harness is in the pages, and they say that Llama Chat does not import @ggml-org/llama-ui yet. Assisted-by: Claude (Opus 5.5 Medium)
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Overview
Llama Chat's agentic loop becomes a headless package of its own, Llama Harness (
@ggml-org/llama-harness), so the loop can run without the UI and an app can use it or bring its own. Llama Chat works as before.Additional information
Part 5 of 13 of the work-in-progress stack on the fork, based on
allozaur/feat/voice-conversation(#65), voice mode, which sits on voice messages and dictation (#64) and the npx package (#69). Its diff is only this layer.Since 17 November this layer sits above the voice PRs (#64, #65), right under the MCP extension (#61); before that it sat below them, and on 13 November it moved down from the top of the stack. Its commits are the same apart from the Projects tools, which the agentic store now gains in the Projects layer above.
The package is at
ui/lib/harness, beside the library inui/lib/ui:ui/libholds the packages, one folder each.Known gaps: the Docker and Nix builds were not run locally. The harness is not published to npm yet, and Llama Chat still uses its own chat service as the inference client, not the package's.
What changed
ui/lib/harnessholds@ggml-org/llama-harness: plain TypeScript, no UI, no DOM, no dependencies.createHarness(client).run(messages, tools, signal, options)returns the run as an async stream of events:turn_start,text,reasoning,tool_call_delta,timings,message,tool_call,tool_update,tool_result,turn_end, thendoneorerror. It never throws, and it waits for the consumer between events, so an app can store each message before the loop goes on.execute. The loop keeps no state and stores nothing.signalstops at once,interruptedends the run at the next safe point, andmaxTurnswithonTurnLimitasks before going past the limit.createOpenAIClientis an inference client for any OpenAI-compatible endpoint. It reads llama-server'stimingsandprompt_progresswhen they are there.agenticStoreno longer has the loop. It gives the harnessChatServiceas the inference client and the enabled tools, each behind the permission gate, and it turns the events into the same callbacks as before. Its public API, the gates and the chat store are unchanged.@ggml-org/llama-harnessalias invite.config.ts(a Vite alias plus a TypeScript path), so it needs no build of the package. The builds that copy or isolateui/apps/chatbringui/lib/harnessalong the same way asui/lib/ui: CMake stages its sources next to the app, the Dockerfiles add aCOPY, the Nix derivation takesui/as its source, andsources.cmakewatches the harness sources.Three behaviours differ from before:
Unknown toolerror as its result. Before, a tool that exists but is disabled could still run after the permission prompt.exec_shell_command) is coalesced: when the UI is slower than the stream, it gets the latest output, not every chunk.One existing behaviour is kept as it is and looks like a bug: after Continue at the turn limit, the turn count starts again at 1, so no new assistant message is created for the next turn and it is written into the previous one.
Testing
ui/lib/harness:npm run check,npm test(18 tests of the loop and the client) andnpm run prepack(build and publint) pass.ui/apps/chat:npm run check,npm run lint, the 833 unit tests andnpm run buildpassed on the earlier top of the stack. In the new position, type-check, lint and the 771 unit tests of this layer pass; the build was not run again. The client tests for tool calls and shell streaming pass.agentic-flow.test.ts, runs the flow end to end with the server and the tools mocked: a tool turn, a denied permission, a failing tool, a streaming tool, an image in a tool result, a steering message, a stream error and an abort. The same test passes on the store as it was before this PR.scripts/ui-assets.cmake) stages the app with the harness beside it, builds it with neither package installed and embeds the assets.get_datetimecall shows the permission prompt, runs and the model answers with the result. From Node, the built package runs a tool call and a final answer against the same server, with no browser.Requirements
Claude (Fable 5.2)