@gtrabanco/pi-nan-provider 1.0.1 → 1.2.1

package/scripts/manual-overrides.ts

@@ -58,19 +58,29 @@ export interface ManualModelOverride {
  * and appear in GeneratedModelEntry.reasoningEffortValues, where the pi
  * model-selector can read them to show the correct granularity.
  *
- * Per-model contract from the docs:
+ * Per-model contract from the docs, corrected by live measurement where the
+ * two disagree (maintainer instruction: the measurement wins over the docs):
  * - glm5.3, glm5.3-flash: fully controllable (low/medium/high/max)
  *   + minimal also accepted (issue #16, measured 2026-09-27): minimal
  *     yields 38 reasoning tokens vs 2,288 for high — the docs simply
  *     omitted it because it was not originally advertised, but it IS
  *     accepted and effective by the gateway.
  * - qwen3.6, gemma4: none/minimal skip reasoning, others cap depth
- * - deepseek-v4-flash: any value accepted but only "none" deterministically
- *   disables reasoning (issue #16, measured 2026-09-27); every other value
- *   (minimal/low/medium/high/omitted) leaves 13-14K reasoning tokens with
- *   runaway reasoning and finish_reason:"length" with ZERO answer. Since only
- *   "none" has a measurable effect, the catalog declares ["none"].
- * - qwen3.8-flash, mimo-v2.6-flash: accepted, depth not adjustable
+ *   (2,048 / 8,192 / 16,384 / 32,768).
+ * - deepseek-v4-flash: `none` reproducibly disables reasoning (0 reasoning
+ *   tokens, full answer); every other value — `minimal`, `low`, `high` and
+ *   omitting the parameter — leaves it ON (issue #16, 2026-09-27,
+ *   re-confirmed live 2026-10-14: none 0 / low 57 / omitted 63 reasoning
+ *   tokens). The docs say "any value (no effect)"; the measurement wins, so
+ *   the catalog declares only ["none"].
+ * - mimo-v2.6-flash: the docs list it as "accepted, depth not adjustable",
+ *   but the live re-probe (2026-10-14) shows `none` → 0 reasoning tokens
+ *   while `low` → 22 and omission → 17. Adjustable at the off end, so it
+ *   carries `thinkingLevelMap { off: "none" }`.
+ * - qwen3.8-flash: the docs say "accepted, depth not adjustable"; the live
+ *   probe measured 0 reasoning tokens for `none` (2026-10-14) but no other
+ *   value was sampled, so no `thinkingLevelMap` is declared — the docs state
+ *   its default is reasoning ON, so `off` sends nothing and the model decides.
  *
  * (mimo-v2.5, which used to share that last row, was removed by NaN — absent
  * from the docs and from openapi.json, checked 2026-09-29.)
@@ -91,7 +101,7 @@ export const REASONING_EFFORT_VALUES: Record<string, string[]> = {
  * sides assert the exact same string.
  */
 const MANUAL_ONLY_NOTE =
-	"omnimodal model (text, image, audio input) served by NaN; limits and modalities from https://nan.builders/docs/models#mimo-v2-6-flash (checked 2026-09-25, quotas re-checked 2026-09-29): 1,048,576 context / 131,072 max output / 1.0B monthly quota per member. Reasoning: accepted, depth not adjustable (docs table, checked 2026-09-29). Tool calling: yes. Streaming: yes. Input modalities (pi-representable): text, image (audio not representable in pi's Model type). Fallback entry: emitted only while models.dev provider nan does not list the model — models.dev started listing it on 2026-09-29, so the generator then keeps the models.dev values and attaches this note to that entry instead of emitting a second one.";
+	"omnimodal model (text, image, audio input) served by NaN; limits and modalities from https://nan.builders/docs/models#mimo-v2-6-flash (checked 2026-09-25, quotas re-checked 2026-09-29): 1,048,576 context / 131,072 max output / 1.0B monthly quota per member. Tool calling: yes. Streaming: yes. Input modalities (pi-representable): text, image (audio not representable in pi's Model type). Fallback entry: emitted only while models.dev provider nan does not list the model — models.dev started listing it on 2026-09-29, so the generator then keeps the models.dev values and attaches this note to that entry instead of emitting a second one.";
 
 /**
  * Models NaN serves that models.dev provider nan may not list. Mirrors the
@@ -139,26 +149,40 @@ export const MANUAL_ONLY_MODELS: Record<string, ManualModelOverride> = {
  *   docs are explicit about both the default and the `none` behavior.
  */
 export const MANUAL_OVERRIDES: Record<string, ManualModelOverride> = {
+	"qwen3.8-flash": {
+		contextWindow: 1_048_576,
+		note:
+			"contextWindow 1,048,576: NaN docs state '1M token context (1,048,576 tokens)' and 'Max answer 131K tokens' (https://nan.builders/docs/models, checked 2026-10-14) — models.dev still lists 262,144. The earlier override was withdrawn 2026-09-07 when the docs read '262K token context, the model's native window'; the docs now state 1M and they are the source of truth (maintainer instruction: docs win over models.dev, which a user updates later). maxTokens stays 131,072, which models.dev already matches and the docs confirm as the 131K max answer.",
+	},
 	"deepseek-v4-flash": {
 		input: ["text", "image"],
+		contextWindow: 1_048_576,
 		maxTokens: 32_768,
 		reasoningEffortValues: ["none"],
 		thinkingLevelMap: { off: "none" },
 		note:
-			"input includes image: NaN serves the Vision-Exp variant ('takes images as input', https://nan.builders/docs/models; the image_url content-parts in https://nan.builders/openapi.json list deepseek-v4-flash among the vision models). models.dev provider nan also lists text+image now (DeepSeek V4.1 Flash entry, checked 2026-09-13; its 2026-09-07 snapshot listed text only), so this override is kept as a pin for the vision capability rather than as a divergence. " +
+			"contextWindow 1,048,576: NaN docs state '1M token context' for this model (https://nan.builders/docs/models, checked 2026-10-14). models.dev lists 1,000,000, a rounded form of the same window; the docs value is authoritative (maintainer instruction: docs win over models.dev, which a user updates later). " +
+			"input includes image: NaN serves the Vision-Exp variant ('takes images as input', https://nan.builders/docs/models; the image_url content-parts in https://nan.builders/openapi.json list deepseek-v4-flash among the vision models). models.dev provider nan also lists text+image now (DeepSeek V4.1 Flash entry, checked 2026-09-13), so this stays a pin for the vision capability rather than a divergence. " +
 			"maxTokens 32768 per the official NaN pi docs (https://nan.builders/docs/pi, 'the same set the NaN CLI writes', checked 2026-10-02) closing issue #20 — the gateway now rejects max_tokens above ~256k for this model with 403 permission_error (measured 2026-10-02: 256,000 → 200; 262,143 → 403), and the previous models.dev-derived 384000 made every pi request fail since pi-ai sends max_tokens = model.maxTokens when no cap is set. models.dev still lists 384000, so this override pins the documented value. " +
 			"reasoning_effort_values=[\"none\"] (issue #16, measured 2026-09-27 against live gateway, repro included in the issue): only \"none\" deterministically disables reasoning (0 reasoning tokens, answer produced). Every other value leaves 13-14K reasoning tokens and causes runaway reasoning (finish_reason:\"length\" with ZERO answer text). The catalog declares only [\"none\"] because the remaining values are useless noise. " +
+			"docs vs measurement (re-probed live 2026-10-14): https://nan.builders/docs/models#controlling-reasoning documents the parameter as 'any value (no effect) — the model decides per request', but the live gateway contradicts that: `none` gave 0 reasoning tokens / 0 reasoning chars / a 134-char answer, while `low` gave 57 reasoning tokens and omitting the parameter gave 63 (`minimal` gave 109). The docs are stale here; the measurement wins (maintainer instruction), so the measured behavior is kept. " +
 			"thinkingLevelMap { off: \"none\" } (issue #16, measured 2026-09-27): when pi maps reasoning:\"off\" to reasoningEffort:undefined, the thinkingLevelMap.off=\"none\" supplies the effort so the gateway receives reasoning_effort:\"none\" and does NOT enter runaway reasoning mode.",
 	},
 	"glm5.3-flash": {
+		contextWindow: 1_048_576,
 		thinkingLevelMap: { off: "minimal" },
 		note:
+			"contextWindow 1,048,576: NaN docs state '1M token context' for this model (https://nan.builders/docs/models, checked 2026-10-14). models.dev lists 1,000,000, a rounded form of the same window; the docs value is authoritative (maintainer instruction: docs win over models.dev, which a user updates later). " +
 			"thinkingLevelMap { off: \"minimal\" } (issue #16, measured 2026-09-27 against live gateway, repro included in the issue): when pi maps reasoning:\"off\" to reasoningEffort:undefined, the thinkingLevelMap.off=\"minimal\" supplies the effort so the gateway receives reasoning_effort:\"minimal\" (38 reasoning tokens + answer). Note: \"none\" does NOT suppress reasoning on glm5.3-flash (13,382 tokens) — only \"minimal\" was measured to be effective. " +
-			"reasoning_effort_values includes \"minimal\" (not documented by NaN docs but accepted and effective by the gateway, issue #16).",
+			"reasoning_effort_values includes \"minimal\" (the docs' glm row lists low/medium/high/max, but the 2026-09-27 measurement shows minimal is accepted and effective, issue #16).",
 	},
-	"qwen3.6": {
+	"mimo-v2.6-flash": {
 		thinkingLevelMap: { off: "none" },
 		note:
+			"thinkingLevelMap { off: \"none\" } (live re-probe 2026-10-14): the docs list this model as 'accepted, depth not adjustable', but the gateway makes the parameter effective at the off end — `none` gave 0 reasoning tokens / a full answer, while `low` gave 22 reasoning tokens and omitting the parameter gave 17. Measured behavior wins over the docs, so `thinking: off` now sends reasoning_effort:\"none\" and actually disables reasoning. No other level was sampled, so only `off` is mapped.",
+	},
+	"qwen3.6": {
+		thinkingLevelMap: { off: "none" },		note:
 			"thinkingLevelMap { off: \"none\" } (issue #16): NaN docs state that with no reasoning_effort parameter every model uses its own default and that default is reasoning ON for qwen3.6 (16,384-token budget), while `none` and `minimal` skip the reasoning phase entirely (https://nan.builders/docs/models#controlling-reasoning, checked 2026-10-01). pi maps reasoning:\"off\" to reasoningEffort:undefined, so without this map `thinking: off` sent no parameter and the model reasoned anyway; the map sends reasoning_effort:\"none\" and actually disables reasoning.",
 	},
 	"gemma4": {

package/scripts/models.generated.ts

@@ -1,7 +1,7 @@
 // This file is auto-generated by scripts/generate-models.ts
 // Do not edit manually — run `bun run generate-models` to update.
 //
-// Source: https://models.dev/api.json (provider "nan"), fetched 2026-10-02T22:53:40.291Z
+// Source: https://models.dev/api.json (provider "nan"), fetched 2026-10-16T17:43:45.590Z
 // Provenance: every contextWindow/maxTokens/input/cost value traces to
 // models.dev or to the per-entry notes below. Nothing is invented; entries
 // models.dev documents incompletely are omitted and flagged instead.
@@ -26,7 +26,7 @@ export const NAN_GENERATED_MODELS: readonly GeneratedModelEntry[] = [
 			"cacheRead": 0,
 			"cacheWrite": 0
 		},
-		"contextWindow": 1000000,
+		"contextWindow": 1048576,
 		"maxTokens": 32768,
 		"reasoningEffortValues": [
 			"none"
@@ -43,7 +43,7 @@ export const NAN_GENERATED_MODELS: readonly GeneratedModelEntry[] = [
 		},
 		"notes": [
 			"compat matches the maintainer's working ~/.pi/agent/models.json LiteLLM config for api.nan.builders (2026-09-04): supportsDeveloperRole false, supportsReasoningEffort true, maxTokensField max_tokens. NaN's docs example sets only supportsDeveloperRole: true and is not battle-tested. supportsFinishReason true (2026-09-13, issue #2): the LiteLLM gateway intermittently closes SSE streams before emitting finish_reason; with true pi-ai raises 'Stream ended without finish_reason', which matches pi-ai's retryable-provider pattern ('ended without') and is retried automatically, whereas false silently synthesized stop/toolUse and stalled the turn mid-answer. supportsUsageInStreaming true (2026-09-16, issue #7): NaN's published schema is silent about stream_options, but the live gateway honors it — two identical streaming calls per model, differing only in stream_options: { include_usage: true }, returned 0 usage chunks without it and exactly 1 with it (prompt/completion/reasoning/cached token counts) on deepseek-v4-flash, glm5.3-flash, qwen3.6, mimo-v2.5 and gemma4, and a real pi session then recorded token counts where it recorded zeros. pi-ai only sends stream_options when this is not false, and the sanitizer forwards it when the model declares true, so chat models opt in by default and usage is reported (issue #4). A model that does not report streaming usage can still opt out per model with a models.json compat override (supportsUsageInStreaming: false); the sanitizer then strips stream_options and the payload stays strict.",
-			"input includes image: NaN serves the Vision-Exp variant ('takes images as input', https://nan.builders/docs/models; the image_url content-parts in https://nan.builders/openapi.json list deepseek-v4-flash among the vision models). models.dev provider nan also lists text+image now (DeepSeek V4.1 Flash entry, checked 2026-09-13; its 2026-09-07 snapshot listed text only), so this override is kept as a pin for the vision capability rather than as a divergence. maxTokens 32768 per the official NaN pi docs (https://nan.builders/docs/pi, 'the same set the NaN CLI writes', checked 2026-10-02) closing issue #20 — the gateway now rejects max_tokens above ~256k for this model with 403 permission_error (measured 2026-10-02: 256,000 → 200; 262,143 → 403), and the previous models.dev-derived 384000 made every pi request fail since pi-ai sends max_tokens = model.maxTokens when no cap is set. models.dev still lists 384000, so this override pins the documented value. reasoning_effort_values=[\"none\"] (issue #16, measured 2026-09-27 against live gateway, repro included in the issue): only \"none\" deterministically disables reasoning (0 reasoning tokens, answer produced). Every other value leaves 13-14K reasoning tokens and causes runaway reasoning (finish_reason:\"length\" with ZERO answer text). The catalog declares only [\"none\"] because the remaining values are useless noise. thinkingLevelMap { off: \"none\" } (issue #16, measured 2026-09-27): when pi maps reasoning:\"off\" to reasoningEffort:undefined, the thinkingLevelMap.off=\"none\" supplies the effort so the gateway receives reasoning_effort:\"none\" and does NOT enter runaway reasoning mode."
+			"contextWindow 1,048,576: NaN docs state '1M token context' for this model (https://nan.builders/docs/models, checked 2026-10-14). models.dev lists 1,000,000, a rounded form of the same window; the docs value is authoritative (maintainer instruction: docs win over models.dev, which a user updates later). input includes image: NaN serves the Vision-Exp variant ('takes images as input', https://nan.builders/docs/models; the image_url content-parts in https://nan.builders/openapi.json list deepseek-v4-flash among the vision models). models.dev provider nan also lists text+image now (DeepSeek V4.1 Flash entry, checked 2026-09-13), so this stays a pin for the vision capability rather than a divergence. maxTokens 32768 per the official NaN pi docs (https://nan.builders/docs/pi, 'the same set the NaN CLI writes', checked 2026-10-02) closing issue #20 — the gateway now rejects max_tokens above ~256k for this model with 403 permission_error (measured 2026-10-02: 256,000 → 200; 262,143 → 403), and the previous models.dev-derived 384000 made every pi request fail since pi-ai sends max_tokens = model.maxTokens when no cap is set. models.dev still lists 384000, so this override pins the documented value. reasoning_effort_values=[\"none\"] (issue #16, measured 2026-09-27 against live gateway, repro included in the issue): only \"none\" deterministically disables reasoning (0 reasoning tokens, answer produced). Every other value leaves 13-14K reasoning tokens and causes runaway reasoning (finish_reason:\"length\" with ZERO answer text). The catalog declares only [\"none\"] because the remaining values are useless noise. docs vs measurement (re-probed live 2026-10-14): https://nan.builders/docs/models#controlling-reasoning documents the parameter as 'any value (no effect) — the model decides per request', but the live gateway contradicts that: `none` gave 0 reasoning tokens / 0 reasoning chars / a 134-char answer, while `low` gave 57 reasoning tokens and omitting the parameter gave 63 (`minimal` gave 109). The docs are stale here; the measurement wins (maintainer instruction), so the measured behavior is kept. thinkingLevelMap { off: \"none\" } (issue #16, measured 2026-09-27): when pi maps reasoning:\"off\" to reasoningEffort:undefined, the thinkingLevelMap.off=\"none\" supplies the effort so the gateway receives reasoning_effort:\"none\" and does NOT enter runaway reasoning mode."
 		],
 		"extras": {
 			"id": "deepseek-v4-flash",
@@ -177,7 +177,7 @@ export const NAN_GENERATED_MODELS: readonly GeneratedModelEntry[] = [
 			"cacheRead": 0,
 			"cacheWrite": 0
 		},
-		"contextWindow": 1000000,
+		"contextWindow": 1048576,
 		"maxTokens": 131072,
 		"reasoningEffortValues": [
 			"minimal",
@@ -198,7 +198,7 @@ export const NAN_GENERATED_MODELS: readonly GeneratedModelEntry[] = [
 		},
 		"notes": [
 			"compat matches the maintainer's working ~/.pi/agent/models.json LiteLLM config for api.nan.builders (2026-09-04): supportsDeveloperRole false, supportsReasoningEffort true, maxTokensField max_tokens. NaN's docs example sets only supportsDeveloperRole: true and is not battle-tested. supportsFinishReason true (2026-09-13, issue #2): the LiteLLM gateway intermittently closes SSE streams before emitting finish_reason; with true pi-ai raises 'Stream ended without finish_reason', which matches pi-ai's retryable-provider pattern ('ended without') and is retried automatically, whereas false silently synthesized stop/toolUse and stalled the turn mid-answer. supportsUsageInStreaming true (2026-09-16, issue #7): NaN's published schema is silent about stream_options, but the live gateway honors it — two identical streaming calls per model, differing only in stream_options: { include_usage: true }, returned 0 usage chunks without it and exactly 1 with it (prompt/completion/reasoning/cached token counts) on deepseek-v4-flash, glm5.3-flash, qwen3.6, mimo-v2.5 and gemma4, and a real pi session then recorded token counts where it recorded zeros. pi-ai only sends stream_options when this is not false, and the sanitizer forwards it when the model declares true, so chat models opt in by default and usage is reported (issue #4). A model that does not report streaming usage can still opt out per model with a models.json compat override (supportsUsageInStreaming: false); the sanitizer then strips stream_options and the payload stays strict.",
-			"thinkingLevelMap { off: \"minimal\" } (issue #16, measured 2026-09-27 against live gateway, repro included in the issue): when pi maps reasoning:\"off\" to reasoningEffort:undefined, the thinkingLevelMap.off=\"minimal\" supplies the effort so the gateway receives reasoning_effort:\"minimal\" (38 reasoning tokens + answer). Note: \"none\" does NOT suppress reasoning on glm5.3-flash (13,382 tokens) — only \"minimal\" was measured to be effective. reasoning_effort_values includes \"minimal\" (not documented by NaN docs but accepted and effective by the gateway, issue #16)."
+			"contextWindow 1,048,576: NaN docs state '1M token context' for this model (https://nan.builders/docs/models, checked 2026-10-14). models.dev lists 1,000,000, a rounded form of the same window; the docs value is authoritative (maintainer instruction: docs win over models.dev, which a user updates later). thinkingLevelMap { off: \"minimal\" } (issue #16, measured 2026-09-27 against live gateway, repro included in the issue): when pi maps reasoning:\"off\" to reasoningEffort:undefined, the thinkingLevelMap.off=\"minimal\" supplies the effort so the gateway receives reasoning_effort:\"minimal\" (38 reasoning tokens + answer). Note: \"none\" does NOT suppress reasoning on glm5.3-flash (13,382 tokens) — only \"minimal\" was measured to be effective. reasoning_effort_values includes \"minimal\" (the docs' glm row lists low/medium/high/max, but the 2026-09-27 measurement shows minimal is accepted and effective, issue #16)."
 		],
 		"extras": {
 			"id": "glm5.3-flash",
@@ -261,6 +261,9 @@ export const NAN_GENERATED_MODELS: readonly GeneratedModelEntry[] = [
 		"contextWindow": 1048576,
 		"maxTokens": 131072,
 		"reasoningEffortValues": [],
+		"thinkingLevelMap": {
+			"off": "none"
+		},
 		"compat": {
 			"supportsDeveloperRole": false,
 			"supportsReasoningEffort": true,
@@ -270,7 +273,8 @@ export const NAN_GENERATED_MODELS: readonly GeneratedModelEntry[] = [
 		},
 		"notes": [
 			"compat matches the maintainer's working ~/.pi/agent/models.json LiteLLM config for api.nan.builders (2026-09-04): supportsDeveloperRole false, supportsReasoningEffort true, maxTokensField max_tokens. NaN's docs example sets only supportsDeveloperRole: true and is not battle-tested. supportsFinishReason true (2026-09-13, issue #2): the LiteLLM gateway intermittently closes SSE streams before emitting finish_reason; with true pi-ai raises 'Stream ended without finish_reason', which matches pi-ai's retryable-provider pattern ('ended without') and is retried automatically, whereas false silently synthesized stop/toolUse and stalled the turn mid-answer. supportsUsageInStreaming true (2026-09-16, issue #7): NaN's published schema is silent about stream_options, but the live gateway honors it — two identical streaming calls per model, differing only in stream_options: { include_usage: true }, returned 0 usage chunks without it and exactly 1 with it (prompt/completion/reasoning/cached token counts) on deepseek-v4-flash, glm5.3-flash, qwen3.6, mimo-v2.5 and gemma4, and a real pi session then recorded token counts where it recorded zeros. pi-ai only sends stream_options when this is not false, and the sanitizer forwards it when the model declares true, so chat models opt in by default and usage is reported (issue #4). A model that does not report streaming usage can still opt out per model with a models.json compat override (supportsUsageInStreaming: false); the sanitizer then strips stream_options and the payload stays strict.",
-			"manual-only: \"mimo-v2.6-flash\" skipped: models.dev provider \"nan\" now lists it, so the manual-only entry is not emitted and the models.dev values are used (omnimodal model (text, image, audio input) served by NaN; limits and modalities from https://nan.builders/docs/models#mimo-v2-6-flash (checked 2026-09-25, quotas re-checked 2026-09-29): 1,048,576 context / 131,072 max output / 1.0B monthly quota per member. Reasoning: accepted, depth not adjustable (docs table, checked 2026-09-29). Tool calling: yes. Streaming: yes. Input modalities (pi-representable): text, image (audio not representable in pi's Model type). Fallback entry: emitted only while models.dev provider nan does not list the model — models.dev started listing it on 2026-09-29, so the generator then keeps the models.dev values and attaches this note to that entry instead of emitting a second one.)"
+			"thinkingLevelMap { off: \"none\" } (live re-probe 2026-10-14): the docs list this model as 'accepted, depth not adjustable', but the gateway makes the parameter effective at the off end — `none` gave 0 reasoning tokens / a full answer, while `low` gave 22 reasoning tokens and omitting the parameter gave 17. Measured behavior wins over the docs, so `thinking: off` now sends reasoning_effort:\"none\" and actually disables reasoning. No other level was sampled, so only `off` is mapped.",
+			"manual-only: \"mimo-v2.6-flash\" skipped: models.dev provider \"nan\" now lists it, so the manual-only entry is not emitted and the models.dev values are used (omnimodal model (text, image, audio input) served by NaN; limits and modalities from https://nan.builders/docs/models#mimo-v2-6-flash (checked 2026-09-25, quotas re-checked 2026-09-29): 1,048,576 context / 131,072 max output / 1.0B monthly quota per member. Tool calling: yes. Streaming: yes. Input modalities (pi-representable): text, image (audio not representable in pi's Model type). Fallback entry: emitted only while models.dev provider nan does not list the model — models.dev started listing it on 2026-09-29, so the generator then keeps the models.dev values and attaches this note to that entry instead of emitting a second one.)"
 		],
 		"extras": {
 			"id": "mimo-v2.6-flash",
@@ -403,7 +407,7 @@ export const NAN_GENERATED_MODELS: readonly GeneratedModelEntry[] = [
 			"cacheRead": 0,
 			"cacheWrite": 0
 		},
-		"contextWindow": 262144,
+		"contextWindow": 1048576,
 		"maxTokens": 131072,
 		"reasoningEffortValues": [],
 		"compat": {
@@ -415,7 +419,7 @@ export const NAN_GENERATED_MODELS: readonly GeneratedModelEntry[] = [
 		},
 		"notes": [
 			"compat matches the maintainer's working ~/.pi/agent/models.json LiteLLM config for api.nan.builders (2026-09-04): supportsDeveloperRole false, supportsReasoningEffort true, maxTokensField max_tokens. NaN's docs example sets only supportsDeveloperRole: true and is not battle-tested. supportsFinishReason true (2026-09-13, issue #2): the LiteLLM gateway intermittently closes SSE streams before emitting finish_reason; with true pi-ai raises 'Stream ended without finish_reason', which matches pi-ai's retryable-provider pattern ('ended without') and is retried automatically, whereas false silently synthesized stop/toolUse and stalled the turn mid-answer. supportsUsageInStreaming true (2026-09-16, issue #7): NaN's published schema is silent about stream_options, but the live gateway honors it — two identical streaming calls per model, differing only in stream_options: { include_usage: true }, returned 0 usage chunks without it and exactly 1 with it (prompt/completion/reasoning/cached token counts) on deepseek-v4-flash, glm5.3-flash, qwen3.6, mimo-v2.5 and gemma4, and a real pi session then recorded token counts where it recorded zeros. pi-ai only sends stream_options when this is not false, and the sanitizer forwards it when the model declares true, so chat models opt in by default and usage is reported (issue #4). A model that does not report streaming usage can still opt out per model with a models.json compat override (supportsUsageInStreaming: false); the sanitizer then strips stream_options and the payload stays strict.",
-			"contextWindow 262,144: the earlier 1,000,000 override (maintainer-confirmed 2026-09-05) was withdrawn 2026-09-07 — the updated https://nan.builders/docs/models still states '262K token context, the model's native window' and models.dev agrees at 262,144; NaN docs are treated as the most reliable source (maintainer instruction, 2026-09-07)."
+			"contextWindow 1,048,576: NaN docs state '1M token context (1,048,576 tokens)' and 'Max answer 131K tokens' (https://nan.builders/docs/models, checked 2026-10-14) — models.dev still lists 262,144. The earlier override was withdrawn 2026-09-07 when the docs read '262K token context, the model's native window'; the docs now state 1M and they are the source of truth (maintainer instruction: docs win over models.dev, which a user updates later). maxTokens stays 131,072, which models.dev already matches and the docs confirm as the 131K max answer."
 		],
 		"extras": {
 			"id": "qwen3.8-flash",
@@ -455,7 +459,7 @@ export const NAN_GENERATED_MODELS: readonly GeneratedModelEntry[] = [
 export const GENERATED_CATALOG_META = {
 	source: "https://models.dev/api.json",
 	modelsDevProvider: "nan",
-	fetchedAt: "2026-10-02T22:53:40.291Z",
+	fetchedAt: "2026-10-16T17:43:45.590Z",
 	modelCount: 6,
 	models: ["deepseek-v4-flash","gemma4","glm5.3-flash","mimo-v2.6-flash","qwen3.6","qwen3.8-flash"],
 	notes: [
@@ -469,11 +473,12 @@ export const GENERATED_CATALOG_META = {
 	"non-chat: \"whisper\" excluded from the chat catalog (speech-to-text (STT) endpoint (https://nan.builders/docs/models, checked 2026-10-01) — not a chat model)",
 	"non-chat: \"flux-2-klein\" excluded from the chat catalog (image generation endpoint (https://nan.builders/docs/models, checked 2026-10-01) — not a chat model)",
 	"compat matches the maintainer's working ~/.pi/agent/models.json LiteLLM config for api.nan.builders (2026-09-04): supportsDeveloperRole false, supportsReasoningEffort true, maxTokensField max_tokens. NaN's docs example sets only supportsDeveloperRole: true and is not battle-tested. supportsFinishReason true (2026-09-13, issue #2): the LiteLLM gateway intermittently closes SSE streams before emitting finish_reason; with true pi-ai raises 'Stream ended without finish_reason', which matches pi-ai's retryable-provider pattern ('ended without') and is retried automatically, whereas false silently synthesized stop/toolUse and stalled the turn mid-answer. supportsUsageInStreaming true (2026-09-16, issue #7): NaN's published schema is silent about stream_options, but the live gateway honors it — two identical streaming calls per model, differing only in stream_options: { include_usage: true }, returned 0 usage chunks without it and exactly 1 with it (prompt/completion/reasoning/cached token counts) on deepseek-v4-flash, glm5.3-flash, qwen3.6, mimo-v2.5 and gemma4, and a real pi session then recorded token counts where it recorded zeros. pi-ai only sends stream_options when this is not false, and the sanitizer forwards it when the model declares true, so chat models opt in by default and usage is reported (issue #4). A model that does not report streaming usage can still opt out per model with a models.json compat override (supportsUsageInStreaming: false); the sanitizer then strips stream_options and the payload stays strict.",
-	"input includes image: NaN serves the Vision-Exp variant ('takes images as input', https://nan.builders/docs/models; the image_url content-parts in https://nan.builders/openapi.json list deepseek-v4-flash among the vision models). models.dev provider nan also lists text+image now (DeepSeek V4.1 Flash entry, checked 2026-09-13; its 2026-09-07 snapshot listed text only), so this override is kept as a pin for the vision capability rather than as a divergence. maxTokens 32768 per the official NaN pi docs (https://nan.builders/docs/pi, 'the same set the NaN CLI writes', checked 2026-10-02) closing issue #20 — the gateway now rejects max_tokens above ~256k for this model with 403 permission_error (measured 2026-10-02: 256,000 → 200; 262,143 → 403), and the previous models.dev-derived 384000 made every pi request fail since pi-ai sends max_tokens = model.maxTokens when no cap is set. models.dev still lists 384000, so this override pins the documented value. reasoning_effort_values=[\"none\"] (issue #16, measured 2026-09-27 against live gateway, repro included in the issue): only \"none\" deterministically disables reasoning (0 reasoning tokens, answer produced). Every other value leaves 13-14K reasoning tokens and causes runaway reasoning (finish_reason:\"length\" with ZERO answer text). The catalog declares only [\"none\"] because the remaining values are useless noise. thinkingLevelMap { off: \"none\" } (issue #16, measured 2026-09-27): when pi maps reasoning:\"off\" to reasoningEffort:undefined, the thinkingLevelMap.off=\"none\" supplies the effort so the gateway receives reasoning_effort:\"none\" and does NOT enter runaway reasoning mode.",
+	"contextWindow 1,048,576: NaN docs state '1M token context' for this model (https://nan.builders/docs/models, checked 2026-10-14). models.dev lists 1,000,000, a rounded form of the same window; the docs value is authoritative (maintainer instruction: docs win over models.dev, which a user updates later). input includes image: NaN serves the Vision-Exp variant ('takes images as input', https://nan.builders/docs/models; the image_url content-parts in https://nan.builders/openapi.json list deepseek-v4-flash among the vision models). models.dev provider nan also lists text+image now (DeepSeek V4.1 Flash entry, checked 2026-09-13), so this stays a pin for the vision capability rather than a divergence. maxTokens 32768 per the official NaN pi docs (https://nan.builders/docs/pi, 'the same set the NaN CLI writes', checked 2026-10-02) closing issue #20 — the gateway now rejects max_tokens above ~256k for this model with 403 permission_error (measured 2026-10-02: 256,000 → 200; 262,143 → 403), and the previous models.dev-derived 384000 made every pi request fail since pi-ai sends max_tokens = model.maxTokens when no cap is set. models.dev still lists 384000, so this override pins the documented value. reasoning_effort_values=[\"none\"] (issue #16, measured 2026-09-27 against live gateway, repro included in the issue): only \"none\" deterministically disables reasoning (0 reasoning tokens, answer produced). Every other value leaves 13-14K reasoning tokens and causes runaway reasoning (finish_reason:\"length\" with ZERO answer text). The catalog declares only [\"none\"] because the remaining values are useless noise. docs vs measurement (re-probed live 2026-10-14): https://nan.builders/docs/models#controlling-reasoning documents the parameter as 'any value (no effect) — the model decides per request', but the live gateway contradicts that: `none` gave 0 reasoning tokens / 0 reasoning chars / a 134-char answer, while `low` gave 57 reasoning tokens and omitting the parameter gave 63 (`minimal` gave 109). The docs are stale here; the measurement wins (maintainer instruction), so the measured behavior is kept. thinkingLevelMap { off: \"none\" } (issue #16, measured 2026-09-27): when pi maps reasoning:\"off\" to reasoningEffort:undefined, the thinkingLevelMap.off=\"none\" supplies the effort so the gateway receives reasoning_effort:\"none\" and does NOT enter runaway reasoning mode.",
 	"thinkingLevelMap { off: \"none\" } (issue #16): NaN docs state that with no reasoning_effort parameter gemma4 defaults to reasoning ON (16,384-token budget), while `none`/`minimal` skip reasoning entirely (https://nan.builders/docs/models#controlling-reasoning, checked 2026-10-01). pi maps reasoning:\"off\" to reasoningEffort:undefined, so without this map `thinking: off` sent no parameter and the model reasoned anyway; the map sends reasoning_effort:\"none\" and actually disables reasoning.",
-	"thinkingLevelMap { off: \"minimal\" } (issue #16, measured 2026-09-27 against live gateway, repro included in the issue): when pi maps reasoning:\"off\" to reasoningEffort:undefined, the thinkingLevelMap.off=\"minimal\" supplies the effort so the gateway receives reasoning_effort:\"minimal\" (38 reasoning tokens + answer). Note: \"none\" does NOT suppress reasoning on glm5.3-flash (13,382 tokens) — only \"minimal\" was measured to be effective. reasoning_effort_values includes \"minimal\" (not documented by NaN docs but accepted and effective by the gateway, issue #16).",
-	"manual-only: \"mimo-v2.6-flash\" skipped: models.dev provider \"nan\" now lists it, so the manual-only entry is not emitted and the models.dev values are used (omnimodal model (text, image, audio input) served by NaN; limits and modalities from https://nan.builders/docs/models#mimo-v2-6-flash (checked 2026-09-25, quotas re-checked 2026-09-29): 1,048,576 context / 131,072 max output / 1.0B monthly quota per member. Reasoning: accepted, depth not adjustable (docs table, checked 2026-09-29). Tool calling: yes. Streaming: yes. Input modalities (pi-representable): text, image (audio not representable in pi's Model type). Fallback entry: emitted only while models.dev provider nan does not list the model — models.dev started listing it on 2026-09-29, so the generator then keeps the models.dev values and attaches this note to that entry instead of emitting a second one.)",
+	"contextWindow 1,048,576: NaN docs state '1M token context' for this model (https://nan.builders/docs/models, checked 2026-10-14). models.dev lists 1,000,000, a rounded form of the same window; the docs value is authoritative (maintainer instruction: docs win over models.dev, which a user updates later). thinkingLevelMap { off: \"minimal\" } (issue #16, measured 2026-09-27 against live gateway, repro included in the issue): when pi maps reasoning:\"off\" to reasoningEffort:undefined, the thinkingLevelMap.off=\"minimal\" supplies the effort so the gateway receives reasoning_effort:\"minimal\" (38 reasoning tokens + answer). Note: \"none\" does NOT suppress reasoning on glm5.3-flash (13,382 tokens) — only \"minimal\" was measured to be effective. reasoning_effort_values includes \"minimal\" (the docs' glm row lists low/medium/high/max, but the 2026-09-27 measurement shows minimal is accepted and effective, issue #16).",
+	"thinkingLevelMap { off: \"none\" } (live re-probe 2026-10-14): the docs list this model as 'accepted, depth not adjustable', but the gateway makes the parameter effective at the off end — `none` gave 0 reasoning tokens / a full answer, while `low` gave 22 reasoning tokens and omitting the parameter gave 17. Measured behavior wins over the docs, so `thinking: off` now sends reasoning_effort:\"none\" and actually disables reasoning. No other level was sampled, so only `off` is mapped.",
+	"manual-only: \"mimo-v2.6-flash\" skipped: models.dev provider \"nan\" now lists it, so the manual-only entry is not emitted and the models.dev values are used (omnimodal model (text, image, audio input) served by NaN; limits and modalities from https://nan.builders/docs/models#mimo-v2-6-flash (checked 2026-09-25, quotas re-checked 2026-09-29): 1,048,576 context / 131,072 max output / 1.0B monthly quota per member. Tool calling: yes. Streaming: yes. Input modalities (pi-representable): text, image (audio not representable in pi's Model type). Fallback entry: emitted only while models.dev provider nan does not list the model — models.dev started listing it on 2026-09-29, so the generator then keeps the models.dev values and attaches this note to that entry instead of emitting a second one.)",
 	"thinkingLevelMap { off: \"none\" } (issue #16): NaN docs state that with no reasoning_effort parameter every model uses its own default and that default is reasoning ON for qwen3.6 (16,384-token budget), while `none` and `minimal` skip the reasoning phase entirely (https://nan.builders/docs/models#controlling-reasoning, checked 2026-10-01). pi maps reasoning:\"off\" to reasoningEffort:undefined, so without this map `thinking: off` sent no parameter and the model reasoned anyway; the map sends reasoning_effort:\"none\" and actually disables reasoning.",
-	"contextWindow 262,144: the earlier 1,000,000 override (maintainer-confirmed 2026-09-05) was withdrawn 2026-09-07 — the updated https://nan.builders/docs/models still states '262K token context, the model's native window' and models.dev agrees at 262,144; NaN docs are treated as the most reliable source (maintainer instruction, 2026-09-07)."
+	"contextWindow 1,048,576: NaN docs state '1M token context (1,048,576 tokens)' and 'Max answer 131K tokens' (https://nan.builders/docs/models, checked 2026-10-14) — models.dev still lists 262,144. The earlier override was withdrawn 2026-09-07 when the docs read '262K token context, the model's native window'; the docs now state 1M and they are the source of truth (maintainer instruction: docs win over models.dev, which a user updates later). maxTokens stays 131,072, which models.dev already matches and the docs confirm as the 131K max answer."
 ],
 } as const;

package/src/commands.ts

@@ -1,24 +1,19 @@
 /**
- * `/nan-mcp` — slash command configuring both native MCP servers this
+ * `/nan-mcp` — slash command configuring the native MCP server this
  * package registers:
  *
- * - `nan-search` — the official NaN remote MCP server (api.nan.builders/mcp),
- *   exposing `mcp__nan_search__web_search` (pi >=0.99.2). Default: enabled.
- *   (pi 0.99.0–0.99.1 used the hyphenated form `mcp__nan-search__web_search`.)
  * - `nan-media` — the community stdio media server (flux-2-klein / kokoro /
  *   whisper), exposing `mcp__nan_media__generate_image` etc. Default: enabled.
  *   (pi 0.99.0–0.99.1 used the hyphenated form `mcp__nan-media__*`.)
  *
- * Both are session-scoped native servers (visible in /mcp, source "extension").
- * /nan-mcp enables/disables them via pi.registerMcpServer / pi.unregisterMcpServer,
+ * The server is session-scoped (visible in /mcp, source "extension").
+ * /nan-mcp enables/disables it via pi.registerMcpServer / pi.unregisterMcpServer,
  * persisting the toggle in nan-provider.json.
  *
- * Targets accept aliases: `web-search` (`search`, `web_search`, `official`,
- * `nan-web-search`) and `nan-mcp-server` (`media`, `media-mcp`, `nan-media`).
- * Explicit env vars (NAN_MCP_TOOLS, NAN_MEDIA_MCP) override the persisted
- * toggles for the session.
+ * Target aliases: `nan-mcp-server` (`media`, `media-mcp`, `nan-media`).
+ * Explicit env var (NAN_MEDIA_MCP) overrides the persisted toggle for the session.
  *
- * Native behavior changes from the old bridge:
+ * Native behavior:
  * - `disable` IMMEDIATELY hides tools (unregisterMcpServer disconnects the
  *   server; the MCP extension hides tools when disconnected).
  * - `enable` IMMEDIATELY registers the server for the current session.
@@ -29,55 +24,31 @@
 
 import type { ExtensionAPI, ExtensionCommandContext } from "@earendil-works/pi-coding-agent";
 import { mediaMcpCommand, mediaMcpEnabled, mediaMcpSource, mediaMcpEnvExplicit, mediaMcpEnvTruthy, mediaMcpTimeoutSec } from "./mcp/media-server.ts";
-import { webSearchBridgeEnabled, webSearchBridgeSource, mcpToolsEnvExplicit, mcpToolsDisabled } from "./mcp/api-key.ts";
 import { NAN_API_KEY_ENV, tryResolveNanApiKeyViaRegistry, resolveStoredNanApiKey } from "./mcp/api-key.ts";
 import { NAN_STATE_FILE, MCP_SERVER_NAMES, readBridgeState, writeBridgeState, type BridgeKey, readState, writeState } from "./mcp/state.ts";
 import { connectorMissingMessage, hasMcpConnector } from "./mcp/host-support.ts";
 
 const USAGE =
-	"Usage: /nan-mcp [status] · /nan-mcp enable [web-search|nan-mcp-server] · /nan-mcp disable [web-search|nan-mcp-server]";
+	"Usage: /nan-mcp [status] · /nan-mcp enable [nan-mcp-server] · /nan-mcp disable [nan-mcp-server]";
 
 /** Canonical bridge targets with aliases. */
 const TARGETS: Record<string, BridgeKey> = {
-	"web-search": "webSearch",
-	search: "webSearch",
-	web_search: "webSearch",
-	official: "webSearch",
-	"nan-web-search": "webSearch",
 	"nan-mcp-server": "mediaMcp",
 	media: "mediaMcp",
 	"media-mcp": "mediaMcp",
 	"nan-media": "mediaMcp",
 };
 
-function targetName(bridge: BridgeKey): string {
-	return bridge === "webSearch" ? "web-search" : "nan-mcp-server";
+function envOverrideLabel(): string {
+	return `env NAN_MEDIA_MCP=${process.env.NAN_MEDIA_MCP ?? ""} (overrides persisted)`;
 }
 
-function envOverrideLabel(bridge: BridgeKey): string {
-	return bridge === "webSearch"
-		? `env NAN_MCP_TOOLS=${process.env.NAN_MCP_TOOLS ?? ""} (overrides persisted)`
-		: `env NAN_MEDIA_MCP=${process.env.NAN_MEDIA_MCP ?? ""} (overrides persisted)`;
-}
-
-function sourceLabel(bridge: BridgeKey, source: "env" | "persisted" | "default"): string {
-	if (source === "env") return envOverrideLabel(bridge);
+function sourceLabel(source: "env" | "persisted" | "default"): string {
+	if (source === "env") return envOverrideLabel();
 	if (source === "persisted") {
-		return `persisted in <agentDir>/${NAN_STATE_FILE} (${targetName(bridge)}: ${readBridgeState(bridge)})`;
+		return `persisted in <agentDir>/${NAN_STATE_FILE} (nan-mcp-server: ${readBridgeState("mediaMcp")})`;
 	}
-	return "default (both bridges are enabled by default)";
-}
-
-/** Resolve a bridge key to its native server name. */
-function serverName(bridge: BridgeKey): string {
-	return MCP_SERVER_NAMES[bridge];
-}
-
-/** Describe the bridge in the status message. */
-function bridgeDescription(bridge: BridgeKey): string {
-	return bridge === "webSearch"
-		? "web-search bridge (official NaN MCP → mcp__nan_search__web_search)"
-		: "nan-mcp-server bridge (community media MCP → mcp__nan_media__generate_image/edit_image/text_to_speech/list_voices/speech_to_text)";
+	return "default (bridge is enabled by default)";
 }
 
 /** Check if a native server of this name is currently registered. */
@@ -89,39 +60,27 @@ function nativeServerRegistered(pi: ExtensionAPI, name: string): boolean {
 	}
 }
 
-/** Status message listing both bridges' gate state and native registration.
+/** Status message listing the media bridge gate state and native registration.
  * `ctx` is the command context (provides modelRegistry for key lookup).
  */
 async function statusMessage(pi: ExtensionAPI, ctx: ExtensionCommandContext): Promise<string> {
 	const lines: string[] = [];
 
 	if (!hasMcpConnector(pi)) {
-		lines.push(connectorMissingMessage([MCP_SERVER_NAMES.webSearch, MCP_SERVER_NAMES.mediaMcp]));
+		lines.push(connectorMissingMessage([MCP_SERVER_NAMES.mediaMcp]));
 	}
 
-	for (const bridge of ["webSearch" as BridgeKey, "mediaMcp" as BridgeKey]) {
-		const enabled = bridge === "webSearch" ? webSearchBridgeEnabled() : mediaMcpEnabled();
-		const source = bridge === "webSearch" ? webSearchBridgeSource() : mediaMcpSource();
-		const regStatus = nativeServerRegistered(pi, serverName(bridge)) ? "registered (native)" : "not registered";
-		lines.push(`${bridgeDescription(bridge)}: ${enabled ? "enabled" : "disabled"} — ${sourceLabel(bridge, source)}, ${regStatus}.`);
-
-		if (bridge === "webSearch") {
-			// Key status for web-search
-			const keyStatus = await resolveKeyStatus(ctx);
-			lines.push(`  API key: ${keyStatus}.`);
-		}
-		if (bridge === "mediaMcp") {
-			lines.push(`  Media spawn command: ${mediaMcpCommand().join(" ")}.`);
-		}
-	}
-
-	lines.push("Configure with /nan-mcp enable|disable [web-search|nan-mcp-server]; no target = both.");
+	const enabled = mediaMcpEnabled();
+	const source = mediaMcpSource();
+	const regStatus = nativeServerRegistered(pi, MCP_SERVER_NAMES.mediaMcp) ? "registered (native)" : "not registered";
+	lines.push(`nan-mcp-server bridge: ${enabled ? "enabled" : "disabled"} — ${sourceLabel(source)}, ${regStatus}.`);
+	lines.push(`  Media spawn command: ${mediaMcpCommand().join(" ")}.`);
+	lines.push(`  API key: ${await resolveKeyStatus(ctx)}.`);
+	lines.push("Configure with /nan-mcp enable|disable [nan-mcp-server]; no target for the media server.");
 	return lines.join("\n");
 }
 
-/** Check API key resolution status for web-search.
- * `ctx` is the command context (provides modelRegistry).
- */
+/** Check API key resolution status. */
 async function resolveKeyStatus(ctx: ExtensionCommandContext): Promise<string> {
 	const registryKey = await tryResolveNanApiKeyViaRegistry(ctx.modelRegistry);
 	const storedKey = resolveStoredNanApiKey();
@@ -130,37 +89,22 @@ async function resolveKeyStatus(ctx: ExtensionCommandContext): Promise<string> {
 	return "NOT SET — run /login nan or export NAN_API_KEY";
 }
 
-/** Parse an optional target argument; no target = both bridges. */
-function parseTarget(token: string | undefined): { bridges: BridgeKey[]; target?: string } | undefined {
-	if (token === undefined) return { bridges: ["webSearch", "mediaMcp"] };
-	const key = TARGETS[token.toLowerCase()];
-	if (!key) return undefined;
-	return { bridges: [key], target: token.toLowerCase() };
-}
-
-function describeBridges(bridges: BridgeKey[]): string {
-	const names = bridges.map((bridge) => targetName(bridge));
-	return names.length === 2 ? "both bridges" : names.join(" and ");
-}
-
-/**
- * Check if a bridge is effectively enabled (gate resolved).
- */
-function bridgeEffectivelyEnabled(bridge: BridgeKey): boolean {
-	return bridge === "webSearch" ? webSearchBridgeEnabled() : mediaMcpEnabled();
+/** Parse an optional target argument; no target = media server. */
+function parseTarget(token: string | undefined): string | undefined {
+	if (token === undefined) return "mediaMcp";
+	return TARGETS[token.toLowerCase()];
 }
 
 export function registerNanMcpCommand(pi: ExtensionAPI): void {
 	pi.registerCommand("nan-mcp", {
 		description:
-			"NaN MCP configuration: enable/disable/status for both native servers (official web-search + community media)",
+			"NaN MCP configuration: enable/disable/status for the community media server",
 		getArgumentCompletions: (argumentPrefix: string) => {
 			const prefix = argumentPrefix.trim().toLowerCase();
 			const items = [
-				{ value: "enable", label: "enable", description: "Enable a native server (or both) and persist it" },
-				{ value: "disable", label: "disable", description: "Disable a native server (or both) persistently" },
+				{ value: "enable", label: "enable", description: "Enable the media server and persist it" },
+				{ value: "disable", label: "disable", description: "Disable the media server persistently" },
 				{ value: "status", label: "status", description: "Show current server status and registration" },
-				{ value: "enable web-search", label: "enable web-search", description: "Enable the official NaN web-search server" },
 				{ value: "enable nan-mcp-server", label: "enable nan-mcp-server", description: "Enable the community media server" },
 			];
 			const filtered = items.filter((item) => item.value.startsWith(prefix));
@@ -181,20 +125,17 @@ export function registerNanMcpCommand(pi: ExtensionAPI): void {
 				if (parsed === undefined || rest.length > 0) {
 					ctx.ui.notify(
 						parsed === undefined
-							? `Unknown target "${[rawTarget, ...rest].filter(Boolean).join(" ")}". Targets: web-search, nan-mcp-server (or omit for both).`
+							? `Unknown target "${[rawTarget, ...rest].filter(Boolean).join(" ")}". Target: nan-mcp-server (or omit).`
 							: USAGE,
 						"warning",
 					);
 					return;
 				}
-				const { bridges, target } = parsed;
 				const enabled = subcommand === "enable";
+				const where = rawTarget ? `for "${rawTarget.toLowerCase()}"` : "for the media server";
 
-				for (const bridge of bridges) {
-					writeBridgeState(bridge, enabled);
-				}
+				writeBridgeState("mediaMcp", enabled);
 
-				const where = target ? `for "${target}"` : "for both bridges";
 				const persistence = `persisted in <agentDir>/${NAN_STATE_FILE}`;
 
 				// Without an extension that handles `mcp_servers_change` pi cannot connect the
@@ -203,84 +144,56 @@ export function registerNanMcpCommand(pi: ExtensionAPI): void {
 				// so it applies as soon as a connector (built-in `mcp`) is available.
 				if (enabled && !hasMcpConnector(pi)) {
 					ctx.ui.notify(
-						`Enabled ${describeBridges(bridges)} ${where}, ${persistence} — NOT registered. ` +
-							connectorMissingMessage(bridges.map(serverName)),
+						`Enabled ${where}, ${persistence} — NOT registered. ` +
+							connectorMissingMessage([MCP_SERVER_NAMES.mediaMcp]),
 						"warning",
 					);
 					return;
 				}
 
 				// Immediate native registration/unregistration.
-				for (const bridge of bridges) {
-					const name = serverName(bridge);
-					if (enabled) {
-						// Only register if gate allows it.
-						if (bridgeEffectivelyEnabled(bridge)) {
-							if (bridge === "webSearch") {
-								// Resolve key: stored credential → registry → env.
-								const registryKey = await tryResolveNanApiKeyViaRegistry(ctx.modelRegistry);
-								const storedKey = resolveStoredNanApiKey();
-								const apiKey = storedKey ?? registryKey ?? process.env[NAN_API_KEY_ENV];
-								if (apiKey) {
-									pi.registerMcpServer(name, {
-										type: "http",
-										url: "https://api.nan.builders/mcp",
-										headers: { Authorization: `Bearer ${apiKey}` },
-										exposure: "direct",
-									});
-								} else {
-									ctx.ui.notify(
-										`${NAN_API_KEY_ENV} is not set. Export it or run /login nan.`,
-										"warning",
-									);
-								}
-							} else {
-								// Resolve key for media: stored credential → registry → env.
-								const registryKey = await tryResolveNanApiKeyViaRegistry(ctx.modelRegistry);
-								const storedKey = resolveStoredNanApiKey();
-								const apiKey = storedKey ?? registryKey ?? process.env[NAN_API_KEY_ENV];
-								if (apiKey) {
-									const mediaArgs = mediaMcpCommand();
-									pi.registerMcpServer(name, {
-										type: "stdio",
-										command: mediaArgs[0]!,
-										args: mediaArgs.slice(1),
-										env: { [NAN_API_KEY_ENV]: apiKey },
-										exposure: "direct",
-										timeout: mediaMcpTimeoutSec(),
-									});
-								} else {
-									ctx.ui.notify(
-										`${NAN_API_KEY_ENV} is not set. Export it or run /login nan.`,
-										"warning",
-									);
-								}
-							}
+				const name = MCP_SERVER_NAMES.mediaMcp;
+				if (enabled) {
+					if (mediaMcpEnabled()) {
+						const registryKey = await tryResolveNanApiKeyViaRegistry(ctx.modelRegistry);
+						const storedKey = resolveStoredNanApiKey();
+						const apiKey = storedKey ?? registryKey ?? process.env[NAN_API_KEY_ENV];
+						if (apiKey) {
+							const mediaArgs = mediaMcpCommand();
+							pi.registerMcpServer(name, {
+								type: "stdio",
+								command: mediaArgs[0]!,
+								args: mediaArgs.slice(1),
+								env: { [NAN_API_KEY_ENV]: apiKey },
+								exposure: "direct",
+								timeout: mediaMcpTimeoutSec(),
+							});
+						} else {
+							ctx.ui.notify(
+								`${NAN_API_KEY_ENV} is not set. Export it or run /login nan.`,
+								"warning",
+							);
 						}
-					} else {
-						pi.unregisterMcpServer(name);
 					}
+				} else {
+					pi.unregisterMcpServer(name);
 				}
 
 				if (enabled) {
-					// Only claim success if servers were actually registered.
-					const allRegistered = bridges.every((b: BridgeKey) =>
-						nativeServerRegistered(pi, serverName(b)),
-					);
-					if (allRegistered) {
+					if (nativeServerRegistered(pi, name)) {
 						ctx.ui.notify(
-							`Enabled ${describeBridges(bridges)} ${where}, ${persistence}. Native servers registered for this session.`,
+							`Enabled ${where}, ${persistence}. Native server registered for this session.`,
 							"info",
 						);
 					} else {
 						ctx.ui.notify(
-							`Enabled ${describeBridges(bridges)} ${where}, ${persistence} — API key missing; servers not registered.`,
+							`Enabled ${where}, ${persistence} — API key missing; server not registered.`,
 							"warning",
 						);
 					}
 				} else {
 					ctx.ui.notify(
-						`Disabled ${describeBridges(bridges)} ${where}, ${persistence}. Native servers unregistered — tools hidden immediately.`,
+						`Disabled ${where}, ${persistence}. Native server unregistered — tools hidden immediately.`,
 						"warning",
 					);
 				}