Skip to content

Claude Code. Any Model. The most powerful AI coding agent now speaks every language.

docs: park two release-review findings, and report three catalog rows

Jack Rudenko1f2fed5ef534
diff --git a/ROADMAP.md b/ROADMAP.md
index 6c3d46e2..449cda54 100644
--- a/ROADMAP.md
+++ b/ROADMAP.md
@@ -590,3 +590,19 @@ never `git checkout` or `git stash`, the index is shared with sibling worktrees.
executed the truncated `Write` and returned `InputValidationError`, while the fixed build
executed no tool at all and retried the turn. Method and raw evidence:
`ai-docs/reports/truncated-toolcall-live-verification.md`.
+
+## Picker: the `all-filtered` notice must say "no capability data", not "not chat-capable"
+
+Status: not started. Found in the v10.1.1 release review (merge of the strict chat rule with v10.1.0's picker outcome states).
+
+`buildDiscoveredModelOutcome` (`model-selector.ts`) returns `all-filtered` whenever every served row fails `isReportedChatCapable`, and `DiscoveryNotice.tsx` renders it as "served N models, none of them chat-capable". That wording dates from when an undescribed model counted as chat. Under the strict rule it also fires when nobody DESCRIBED the rows: LM Studio's picker path reads `/v1/models` (no `type` field) while `--probe` reads `/api/v0/models` and its `type`, so the two screens disagree about the same daemon. Hiding the rows is correct; the sentence is false. Carry the undescribed count (`unavailableForMissingCapability` already computes it for the probe path) on the variant and word the notice from it; optionally read LM Studio's `type` in the picker fetcher too.
+
+**Trigger condition**: the next change to `buildDiscoveredModelOutcome` or `DiscoveryNotice.tsx`, or the first user report of a picker notice calling local models not chat-capable.
+
+## Picker tests: cover admission by catalog evidence alone
+
+Status: not started. Found in the same review.
+
+No picker-level test covers an `openai-models-list` provider (13 builtins plus LM Studio), whose discovered rows carry no `reported` and are admitted by the catalog alone; only the negative case is tested. `model-selector-discovery.test.ts` also now uses `mock.module("…/all-models-cache.js")` (contained, restored in `afterAll`), and its STATE C endpoint assertion is circular because the fixture sets the endpoint. Add a positive case (a served id the catalog fixture knows by alias is kept with no `reported`), assert the endpoint on a path the fixture does not control, and replace the module mock with a test-only catalog hook like `catalog-client`'s `_setCatalogEntriesForTest`.
+
+**Trigger condition**: before the next change to the discovered-row admission path, or with the item above.
diff --git a/ai-docs/reports/BACKEND_v1414_contract_verification-20260923.md b/ai-docs/reports/BACKEND_v1414_contract_verification-20260923.md
index 7db5f2cd..93daf65d 100644
--- a/ai-docs/reports/BACKEND_v1414_contract_verification-20260923.md
+++ b/ai-docs/reports/BACKEND_v1414_contract_verification-20260923.md
@@ -296,3 +296,23 @@ publishes no capability field at all, which no backend work reaches: a plain Ope
are in that bucket, and their models are reachable only by explicit `provider@model` until either
the catalog describes them or the endpoint starts publishing. LiteLLM's `/model/info` does
publish a `mode` field; reading it is not yet implemented and would shrink that bucket.
+
+### The rule claudish applies, and three rows it trusts that look wrong
+
+Decided 2026-09-23 and shipped in claudish v10.1.1: a chat model has `"text"` among its
+`inputModalities` AND among its `outputModalities`; other modalities on either side never exclude
+it. An absent, `null` or `[]` input list is read as unknown and denies nothing (your contract).
+So your published modalities are now the whole decision for every catalog row — there is no name
+rule behind them any more — and a wrong row is shown or hidden exactly as published.
+
+Measured on generation `g-20260923031256925-3d26cd04`, three generator models publish
+`in: ["text"]`, `out: ["text"]` and are therefore offered as chat models:
+
+| modelId | published | looks like |
+|---|---|---|
+| `flux-dev-finetuner` | `in: [text]`, `out: [text]` | an image-model fine-tuning job |
+| `flux-fill` | `in: [text]`, `out: [text]` | image inpainting (`out` should include `image`) |
+| `seedance-2.5-el` | `in: [text]`, `out: [text]` | a Seedance video variant (`videoOutput` unset) |
+
+Please check these three. claudish logs them once per process (`[Models] … the catalog publishes
+as chat, from a generator family`) rather than filtering them itself.