Pandore/homarr
Public Access
1
0 Fork 0
Code Actions Activity

feat: add the community-funded Homarr assistant provider (#6563)

Browse Source 
* Add Homarr provider proxy foundation

* Complete Homarr provider integration

* Harden Homarr provider integration

* Address Homarr provider review feedback

* Address CodeRabbit provider feedback

* Localize provider errors

* Handle provider activity load failures

* docs: improve workshop and homepage

* Harden Homarr provider limits

* fix: harden the Homarr provider

* fix: address provider review feedback

* fix: clarify provider request accounting

* feat: complete the Homarr provider endpoint

* fix: terminate invalid provider streams
This commit is contained in:
Thomas authored and ajnart committed 2026-08-14 11:32:11 +02:00
1 parent b8f360c04e
commit 45d8bc9443
77 files changed
+4035 -763
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

apps/docs/docs/management/custom-widgets/agent-authoring.mdx
+17 -4
View File
@@ -125,20 +125,33 @@ To exercise the real Homarr Assistant lifecycle instead of accepting a returned
OPENROUTER_API_KEY=... pnpm --filter @homarr/custom-widgets test:assistant-ai
```
The same suites can target the Homarr provider with a Workshop PocketBase token instead of an OpenRouter key:
```sh
AI_PROVIDER_BASE_URL=https://homarr.dev/api/ai/v1 \
AI_PROVIDER_API_KEY=<workshop-token> \
AI_PROVIDER_MODEL=homarr/model \
pnpm --filter @homarr/custom-widgets test:assistant-ai
```
`AI_PROVIDER_JUDGE_MODEL` can select a separate judge when the configured endpoint offers one. The Homarr provider
offers exactly one model, so generation and review both use `homarr/model`.
The assistant test exposes the same lazy skill, live schema, component catalog, targeted component references,
validation, preview-query, and create-from-preview tools as the product. It refuses to save a definition that was not
validated exactly or whose preview queries were not all tested. Multi-request fixtures return a different payload for
each documented path and require manual-query parameters. Use `--case=pokedex`, `--case=coinmarketcap-keyless`,
`--case=bored-activity`, or `--case=agify-name` for a focused run.
Both suites default generation and harsh review to OpenRouter's `~deepseek/deepseek-v4-flash-latest` route. They save
tool traces, validated manifests, raw reviews, and detailed category scores under
Both suites use the configured provider models for generation and harsh review. They save tool traces, validated
manifests, raw reviews, and detailed category scores under
`packages/custom-widgets/.ai-evaluations/`. These live tests are maintainer-invoked rather than CI-required because
they consume a configured provider account and exercise a nondeterministic model. The repository's ordinary tests
cover the lifecycle enforcement, fixture routing, score calculation, and failure behavior without network access.
Each scenario stops as soon as it passes and has a hard limit of ten generation attempts. Override the models or
per-scenario limit with `OPENROUTER_GENERATOR_MODEL`, `OPENROUTER_JUDGE_MODEL`, and
Each scenario stops as soon as it passes and has a hard limit of ten generation attempts. Override the endpoint,
models, or per-scenario limit with `AI_PROVIDER_BASE_URL`, `AI_PROVIDER_API_KEY`, `AI_PROVIDER_MODEL`,
`AI_PROVIDER_JUDGE_MODEL`, and
`CUSTOM_WIDGET_AI_MAX_LOOPS`. Generated artifacts are local and ignored by Git; promote a reviewed, validated result
into the bundled-widget source deliberately.