oh-my-opencode 5.0.0-beta.85 → 5.0.0-beta.86

package/.agents/skills/codex-qa/SKILL.md

@@ -34,6 +34,8 @@ Verified against `codex-cli 0.140.0` (node, jq, tmux, bun on macOS). Confirm wit
   [references/app-server.md](references/app-server.md).
 - **The captured JSON / pane IS the evidence** — write it under
   `.omo/evidence/<YYYYMMDD>-<slug>/` (no evidence file == the QA did not happen).
+  That directory is gitignored: the files stay local, the PR body carries the
+  summary and decisive excerpts, and nothing under it is ever committed.
 
 ## Setup
 

package/.agents/skills/senpi-qa/SKILL.md

@@ -15,8 +15,13 @@ count as live QA here: `bun run test:senpi` is the package gate, the drivers in
 - **Evidence lives at exactly one path.** Every artifact goes under
   `.omo/evidence/omo-senpi-adapter/<slug>/`. Pick it with
   `scripts/resolve-evidence-dir.mjs` and nothing else — a hand-typed path is how
-  runs end up somewhere like `local-ignore/qa-evidence/`, which no reviewer reads
-  and the PR cannot cite.
+  runs end up somewhere like `local-ignore/qa-evidence/` or a `.qa-evidence/` at
+  the worktree root, which is outside the ignored root and gets committed by
+  accident (#8703).
+- **Evidence stays local.** `.omo/evidence/` is gitignored and the
+  tracked-evidence audit test fails the build if any evidence path is tracked.
+  Never `git add -f` an artifact; the PR body carries the summary and the
+  decisive excerpts.
 - **The real agent dir stays untouched.** The live drivers build their own
   isolated `SENPI_CODING_AGENT_DIR` and deliberately IGNORE a caller-provided
   one, so `~/.senpi/agent` is never used as the sandbox. Report the driver's
@@ -26,7 +31,8 @@ count as live QA here: `bun run test:senpi` is the package gate, the drivers in
   report `SKIP` or `FAIL` in their final JSON rather than degrading to the real
   home. A `SKIP` is not a pass — say so in the evidence README.
 - **The captured JSON is the evidence.** No file on disk means the QA did not
-  happen, which means no commit and no push.
+  happen, which means no commit and no push. The file proves the run on the
+  machine that made it; it is not something the commit carries.
 
 ## Resolve the evidence directory first
 

package/.agents/skills/senpi-qa/scripts/resolve-evidence-dir.mjs

@@ -2,10 +2,10 @@
 // resolve-evidence-dir.mjs - pick the canonical evidence directory for a live Senpi QA run.
 //
 // Live Senpi QA artifacts belong under .omo/evidence/omo-senpi-adapter/<slug>/ and nowhere else.
-// That path is the repository-standard location every reviewer and AGENTS.md rule points at, so a
-// hand-typed alternative (local-ignore/qa-evidence/..., a temp dir, a traversal) silently strands
-// the evidence where the PR cannot cite it. Resolving through this script makes the path a checked
-// contract instead of a convention.
+// That path is the repository-standard location every AGENTS.md rule points at, and it is gitignored,
+// so a hand-typed alternative (local-ignore/qa-evidence/..., .qa-evidence/ at the worktree root, a temp
+// dir, a traversal) either strands the evidence or lands it in a commit (#8703). Resolving through
+// this script makes the path a checked contract instead of a convention.
 //
 // The resolver COMPUTES and VALIDATES only; creating the directory stays the caller's decision, so
 // a rejected slug never leaves a stray directory behind.

package/README.md

@@ -266,7 +266,7 @@ When the main agent delegates to a subagent, it doesn't pick a model. It picks a
 | `quick`              | Single-file changes, typos         |
 | `ultrabrain`         | Hard logic, architecture decisions |
 
-The agent says what kind of work it needs; the harness picks the right model. `ultrabrain` routes to GPT-6 Astra max and `deep` to GPT-6 Astra high (OpenAI / OpenAI Codex, then GitHub Copilot, then OpenCode); both fall back to GPT-5.6 Sol. You touch nothing.
+The agent says what kind of work it needs; the harness picks the right model. `ultrabrain` routes to GPT-6 Astra max (falling back to GPT-5.6 Sol), `deep-high` to GPT-6 Astra xhigh, and `deep-low` to GPT-6 Sol Fast medium (then GPT-6 Sol where the Fast tier is not served). You touch nothing.
 
 ### Claude Code Compatibility
 

package/dist/cli/index.js

@@ -64,7 +64,7 @@ var package_default;
 var init_package = __esm(() => {
   package_default = {
     name: "oh-my-opencode",
-    version: "5.0.0-beta.85",
+    version: "5.0.0-beta.86",
     description: "The Best AI Agent Harness - Batteries-Included OpenCode Plugin with Multi-Model Orchestration, Parallel Background Agents, and Crafted LSP/AST Tools",
     main: "./dist/index.js",
     types: "dist/index.d.ts",
@@ -247,7 +247,7 @@ var init_package = __esm(() => {
       zod: "^4.6.5"
     },
     devDependencies: {
-      "@code-yeongyu/senpi": "2026.9.23",
+      "@code-yeongyu/senpi": "2026.9.23-2",
       "@oh-my-opencode/agents-md-core": "workspace:*",
       "@oh-my-opencode/ast-grep-mcp": "workspace:*",
       "@oh-my-opencode/boulder-state": "workspace:*",
@@ -287,18 +287,18 @@ var init_package = __esm(() => {
       typescript: "^7.0.2"
     },
     optionalDependencies: {
-      "oh-my-opencode-darwin-arm64": "5.0.0-beta.85",
-      "oh-my-opencode-darwin-x64": "5.0.0-beta.85",
-      "oh-my-opencode-darwin-x64-baseline": "5.0.0-beta.85",
-      "oh-my-opencode-linux-arm64": "5.0.0-beta.85",
-      "oh-my-opencode-linux-arm64-musl": "5.0.0-beta.85",
-      "oh-my-opencode-linux-x64": "5.0.0-beta.85",
-      "oh-my-opencode-linux-x64-baseline": "5.0.0-beta.85",
-      "oh-my-opencode-linux-x64-musl": "5.0.0-beta.85",
-      "oh-my-opencode-linux-x64-musl-baseline": "5.0.0-beta.85",
-      "oh-my-opencode-windows-arm64": "5.0.0-beta.85",
-      "oh-my-opencode-windows-x64": "5.0.0-beta.85",
-      "oh-my-opencode-windows-x64-baseline": "5.0.0-beta.85"
+      "oh-my-opencode-darwin-arm64": "5.0.0-beta.86",
+      "oh-my-opencode-darwin-x64": "5.0.0-beta.86",
+      "oh-my-opencode-darwin-x64-baseline": "5.0.0-beta.86",
+      "oh-my-opencode-linux-arm64": "5.0.0-beta.86",
+      "oh-my-opencode-linux-arm64-musl": "5.0.0-beta.86",
+      "oh-my-opencode-linux-x64": "5.0.0-beta.86",
+      "oh-my-opencode-linux-x64-baseline": "5.0.0-beta.86",
+      "oh-my-opencode-linux-x64-musl": "5.0.0-beta.86",
+      "oh-my-opencode-linux-x64-musl-baseline": "5.0.0-beta.86",
+      "oh-my-opencode-windows-arm64": "5.0.0-beta.86",
+      "oh-my-opencode-windows-x64": "5.0.0-beta.86",
+      "oh-my-opencode-windows-x64-baseline": "5.0.0-beta.86"
     },
     overrides: {
       "@earendil-works/pi-agent-core": "0.84.2",
@@ -7234,26 +7234,20 @@ var init_agent_model_requirements = __esm(() => {
       fallbackChain: [
         { providers: ["kimi-for-coding"], model: "kimi-for-coding-highspeed", variant: "off" },
         { providers: ["openai", "chatgpt-subscription"], model: "gpt-6-luna-fast", variant: "low" },
-        { providers: ["deepseek"], model: "deepseek-v4-flash", variant: "max" },
+        { providers: ["deepseek"], model: "deepseek-flash", variant: "max" },
         { providers: ["opencode-go", "bailian-coding-plan"], model: "qwen3.7-plus" },
-        { providers: ["opencode-go"], model: "minimax-m3" },
-        { providers: ["minimax-coding-plan", "minimax-cn-coding-plan"], model: "MiniMax-M3" },
         { providers: ["opencode-go"], model: "minimax-m2.7" },
-        { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" },
-        { providers: ["openai", "chatgpt-subscription"], model: "gpt-5.4-nano" }
+        { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" }
       ]
     },
     explore: {
       fallbackChain: [
         { providers: ["kimi-for-coding"], model: "kimi-for-coding-highspeed", variant: "off" },
         { providers: ["openai", "chatgpt-subscription"], model: "gpt-6-luna-fast", variant: "low" },
-        { providers: ["deepseek"], model: "deepseek-v4-flash", variant: "max" },
+        { providers: ["deepseek"], model: "deepseek-flash", variant: "max" },
         { providers: ["opencode-go", "bailian-coding-plan"], model: "qwen3.7-plus" },
-        { providers: ["opencode-go"], model: "minimax-m3" },
-        { providers: ["minimax-coding-plan", "minimax-cn-coding-plan"], model: "MiniMax-M3" },
         { providers: ["opencode-go"], model: "minimax-m2.7" },
-        { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" },
-        { providers: ["openai", "chatgpt-subscription"], model: "gpt-5.4-nano" }
+        { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" }
       ]
     },
     "multimodal-looker": {
@@ -7387,15 +7381,11 @@ var init_category_model_requirements = __esm(() => {
     },
     "deep-low": {
       fallbackChain: [
+        { providers: ["openai", "chatgpt-subscription"], model: "gpt-6-sol-fast", variant: "medium" },
         {
           providers: ["openai", "chatgpt-subscription", "github-copilot", "opencode"],
           model: "gpt-6-sol",
           variant: "medium"
-        },
-        {
-          providers: ["openai", "chatgpt-subscription", "github-copilot", "opencode"],
-          model: "gpt-5.6-sol",
-          variant: "medium"
         }
       ]
     },
@@ -7404,7 +7394,7 @@ var init_category_model_requirements = __esm(() => {
         {
           providers: ["openai", "chatgpt-subscription", "github-copilot", "opencode"],
           model: "gpt-6-astra",
-          variant: "high"
+          variant: "xhigh"
         }
       ]
     },
@@ -7430,7 +7420,7 @@ var init_category_model_requirements = __esm(() => {
     quick: {
       fallbackChain: [
         { providers: ["chatgpt-subscription"], model: "gpt-6-luna-fast", variant: "low" },
-        { providers: ["deepseek"], model: "deepseek-v4-flash", variant: "off" },
+        { providers: ["deepseek"], model: "deepseek-flash", variant: "off" },
         {
           providers: ["qwen-token-plan", "alibaba-token-plan", "bailian-coding-plan"],
           model: "qwen3.6-flash",
@@ -8175,6 +8165,22 @@ var init_supplemental_entries = __esm(() => {
         output: 128000
       }
     },
+    "gpt-6-sol-fast": {
+      id: "gpt-6-sol-fast",
+      family: "gpt",
+      reasoning: true,
+      temperature: false,
+      toolCall: true,
+      modalities: {
+        input: ["text", "image"],
+        output: ["text"]
+      },
+      limit: {
+        context: 1050000,
+        input: 922000,
+        output: 128000
+      }
+    },
     "gpt-6-luna": {
       id: "gpt-6-luna",
       family: "gpt-nano",
@@ -8255,6 +8261,21 @@ var init_supplemental_entries = __esm(() => {
         output: 128000
       }
     },
+    "deepseek-flash": {
+      id: "deepseek-flash",
+      family: "deepseek-flash",
+      reasoning: true,
+      temperature: true,
+      toolCall: true,
+      modalities: {
+        input: ["text", "image"],
+        output: ["text"]
+      },
+      limit: {
+        context: 1e6,
+        output: 384000
+      }
+    },
     "gpt-6-luna-fast": {
       id: "gpt-6-luna-fast",
       family: "gpt-nano",
@@ -9396,7 +9417,7 @@ You are working on tasks that don't fit specific categories but require substant
 </Category_Context>`, UNSPECIFIED_HIGH_CATEGORY_CALLER_GUIDANCE = `<Selection_Gate>Use only when no specialist category fits and substantial effort spans systems/modules with broad impact. Use unspecified-low for contained moderate work.</Selection_Gate>`, DEEP_LOW_CATEGORY_CALLER_GUIDANCE = `<Selection_Gate>Route here when one subsystem plus its callers holds the mechanism and the evidence, once read, leaves one right answer. Wide but mechanical work belongs here or in a quick batch. When unsure, choose deep-low: a misrouted child returns \`ESCALATE: deep-high\` after one cheap attempt; re-spawn the same brief as deep-high with its findings.</Selection_Gate>`, DEEP_HIGH_CATEGORY_CALLER_GUIDANCE = `<Selection_Gate>Route here only when you can name the decision evidence cannot settle: a trade-off with no single right answer, a contract change crossing a package or process boundary, a mechanism with no in-repo pattern to copy, or correctness argued from invariants rather than observed in a test. Wide scope with easy decisions is deep-low or unspecified-high; reasoning as the deliverable is ultrabrain.</Selection_Gate>`, DEEP_LOW_GATE_MODELS, DEEP_HIGH_GATE_MODEL = "gpt-6-astra", OPENAI_CATEGORIES;
 var init_openai_categories = __esm(() => {
   init_types();
-  DEEP_LOW_GATE_MODELS = ["gpt-6-sol", "gpt-5.6-sol"];
+  DEEP_LOW_GATE_MODELS = ["gpt-6-sol-fast", "gpt-6-sol"];
   OPENAI_CATEGORIES = [
     {
       name: "ultrabrain",
@@ -9407,7 +9428,7 @@ var init_openai_categories = __esm(() => {
     },
     {
       name: "deep-low",
-      config: { model: "openai/gpt-6-sol", variant: "medium" },
+      config: { model: "openai/gpt-6-sol-fast", variant: "medium" },
       description: "Default deep lane: one goal, one deliverable, decisions the child can settle from what it reads. **3D graphics, computer/browser use, CAPTCHA, multimodal, backend, logic, and algorithm work is routed here.** Multiple goals fan out as parallel calls.",
       callerGuidance: DEEP_LOW_CATEGORY_CALLER_GUIDANCE,
       promptAppend: DEEP_LOW_CATEGORY_PROMPT_APPEND,
@@ -9416,7 +9437,7 @@ var init_openai_categories = __esm(() => {
     },
     {
       name: "deep-high",
-      config: { model: "openai/gpt-6-astra", variant: "high" },
+      config: { model: "openai/gpt-6-astra", variant: "xhigh" },
       description: "Escalation deep lane: a goal whose central decision cannot be settled from evidence alone. Same one-goal, one-deliverable contract as deep-low.",
       callerGuidance: DEEP_HIGH_CATEGORY_CALLER_GUIDANCE,
       promptAppend: DEEP_HIGH_CATEGORY_PROMPT_APPEND,
@@ -88732,7 +88753,7 @@ var package_default2;
 var init_package2 = __esm(() => {
   package_default2 = {
     name: "@oh-my-opencode/omo-codex",
-    version: "5.0.0-beta.85",
+    version: "5.0.0-beta.86",
     type: "module",
     private: true,
     description: "Codex harness adapter for oh-my-openagent. Vendored Codex plugin namespace (omo) + TypeScript installer + telemetry.",

package/dist/cli-node/index.js

@@ -65,7 +65,7 @@ var package_default;
 var init_package = __esm(() => {
   package_default = {
     name: "oh-my-opencode",
-    version: "5.0.0-beta.85",
+    version: "5.0.0-beta.86",
     description: "The Best AI Agent Harness - Batteries-Included OpenCode Plugin with Multi-Model Orchestration, Parallel Background Agents, and Crafted LSP/AST Tools",
     main: "./dist/index.js",
     types: "dist/index.d.ts",
@@ -248,7 +248,7 @@ var init_package = __esm(() => {
       zod: "^4.6.5"
     },
     devDependencies: {
-      "@code-yeongyu/senpi": "2026.9.23",
+      "@code-yeongyu/senpi": "2026.9.23-2",
       "@oh-my-opencode/agents-md-core": "workspace:*",
       "@oh-my-opencode/ast-grep-mcp": "workspace:*",
       "@oh-my-opencode/boulder-state": "workspace:*",
@@ -288,18 +288,18 @@ var init_package = __esm(() => {
       typescript: "^7.0.2"
     },
     optionalDependencies: {
-      "oh-my-opencode-darwin-arm64": "5.0.0-beta.85",
-      "oh-my-opencode-darwin-x64": "5.0.0-beta.85",
-      "oh-my-opencode-darwin-x64-baseline": "5.0.0-beta.85",
-      "oh-my-opencode-linux-arm64": "5.0.0-beta.85",
-      "oh-my-opencode-linux-arm64-musl": "5.0.0-beta.85",
-      "oh-my-opencode-linux-x64": "5.0.0-beta.85",
-      "oh-my-opencode-linux-x64-baseline": "5.0.0-beta.85",
-      "oh-my-opencode-linux-x64-musl": "5.0.0-beta.85",
-      "oh-my-opencode-linux-x64-musl-baseline": "5.0.0-beta.85",
-      "oh-my-opencode-windows-arm64": "5.0.0-beta.85",
-      "oh-my-opencode-windows-x64": "5.0.0-beta.85",
-      "oh-my-opencode-windows-x64-baseline": "5.0.0-beta.85"
+      "oh-my-opencode-darwin-arm64": "5.0.0-beta.86",
+      "oh-my-opencode-darwin-x64": "5.0.0-beta.86",
+      "oh-my-opencode-darwin-x64-baseline": "5.0.0-beta.86",
+      "oh-my-opencode-linux-arm64": "5.0.0-beta.86",
+      "oh-my-opencode-linux-arm64-musl": "5.0.0-beta.86",
+      "oh-my-opencode-linux-x64": "5.0.0-beta.86",
+      "oh-my-opencode-linux-x64-baseline": "5.0.0-beta.86",
+      "oh-my-opencode-linux-x64-musl": "5.0.0-beta.86",
+      "oh-my-opencode-linux-x64-musl-baseline": "5.0.0-beta.86",
+      "oh-my-opencode-windows-arm64": "5.0.0-beta.86",
+      "oh-my-opencode-windows-x64": "5.0.0-beta.86",
+      "oh-my-opencode-windows-x64-baseline": "5.0.0-beta.86"
     },
     overrides: {
       "@earendil-works/pi-agent-core": "0.84.2",
@@ -7235,26 +7235,20 @@ var init_agent_model_requirements = __esm(() => {
       fallbackChain: [
         { providers: ["kimi-for-coding"], model: "kimi-for-coding-highspeed", variant: "off" },
         { providers: ["openai", "chatgpt-subscription"], model: "gpt-6-luna-fast", variant: "low" },
-        { providers: ["deepseek"], model: "deepseek-v4-flash", variant: "max" },
+        { providers: ["deepseek"], model: "deepseek-flash", variant: "max" },
         { providers: ["opencode-go", "bailian-coding-plan"], model: "qwen3.7-plus" },
-        { providers: ["opencode-go"], model: "minimax-m3" },
-        { providers: ["minimax-coding-plan", "minimax-cn-coding-plan"], model: "MiniMax-M3" },
         { providers: ["opencode-go"], model: "minimax-m2.7" },
-        { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" },
-        { providers: ["openai", "chatgpt-subscription"], model: "gpt-5.4-nano" }
+        { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" }
       ]
     },
     explore: {
       fallbackChain: [
         { providers: ["kimi-for-coding"], model: "kimi-for-coding-highspeed", variant: "off" },
         { providers: ["openai", "chatgpt-subscription"], model: "gpt-6-luna-fast", variant: "low" },
-        { providers: ["deepseek"], model: "deepseek-v4-flash", variant: "max" },
+        { providers: ["deepseek"], model: "deepseek-flash", variant: "max" },
         { providers: ["opencode-go", "bailian-coding-plan"], model: "qwen3.7-plus" },
-        { providers: ["opencode-go"], model: "minimax-m3" },
-        { providers: ["minimax-coding-plan", "minimax-cn-coding-plan"], model: "MiniMax-M3" },
         { providers: ["opencode-go"], model: "minimax-m2.7" },
-        { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" },
-        { providers: ["openai", "chatgpt-subscription"], model: "gpt-5.4-nano" }
+        { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" }
       ]
     },
     "multimodal-looker": {
@@ -7388,15 +7382,11 @@ var init_category_model_requirements = __esm(() => {
     },
     "deep-low": {
       fallbackChain: [
+        { providers: ["openai", "chatgpt-subscription"], model: "gpt-6-sol-fast", variant: "medium" },
         {
           providers: ["openai", "chatgpt-subscription", "github-copilot", "opencode"],
           model: "gpt-6-sol",
           variant: "medium"
-        },
-        {
-          providers: ["openai", "chatgpt-subscription", "github-copilot", "opencode"],
-          model: "gpt-5.6-sol",
-          variant: "medium"
         }
       ]
     },
@@ -7405,7 +7395,7 @@ var init_category_model_requirements = __esm(() => {
         {
           providers: ["openai", "chatgpt-subscription", "github-copilot", "opencode"],
           model: "gpt-6-astra",
-          variant: "high"
+          variant: "xhigh"
         }
       ]
     },
@@ -7431,7 +7421,7 @@ var init_category_model_requirements = __esm(() => {
     quick: {
       fallbackChain: [
         { providers: ["chatgpt-subscription"], model: "gpt-6-luna-fast", variant: "low" },
-        { providers: ["deepseek"], model: "deepseek-v4-flash", variant: "off" },
+        { providers: ["deepseek"], model: "deepseek-flash", variant: "off" },
         {
           providers: ["qwen-token-plan", "alibaba-token-plan", "bailian-coding-plan"],
           model: "qwen3.6-flash",
@@ -8176,6 +8166,22 @@ var init_supplemental_entries = __esm(() => {
         output: 128000
       }
     },
+    "gpt-6-sol-fast": {
+      id: "gpt-6-sol-fast",
+      family: "gpt",
+      reasoning: true,
+      temperature: false,
+      toolCall: true,
+      modalities: {
+        input: ["text", "image"],
+        output: ["text"]
+      },
+      limit: {
+        context: 1050000,
+        input: 922000,
+        output: 128000
+      }
+    },
     "gpt-6-luna": {
       id: "gpt-6-luna",
       family: "gpt-nano",
@@ -8256,6 +8262,21 @@ var init_supplemental_entries = __esm(() => {
         output: 128000
       }
     },
+    "deepseek-flash": {
+      id: "deepseek-flash",
+      family: "deepseek-flash",
+      reasoning: true,
+      temperature: true,
+      toolCall: true,
+      modalities: {
+        input: ["text", "image"],
+        output: ["text"]
+      },
+      limit: {
+        context: 1e6,
+        output: 384000
+      }
+    },
     "gpt-6-luna-fast": {
       id: "gpt-6-luna-fast",
       family: "gpt-nano",
@@ -9397,7 +9418,7 @@ You are working on tasks that don't fit specific categories but require substant
 </Category_Context>`, UNSPECIFIED_HIGH_CATEGORY_CALLER_GUIDANCE = `<Selection_Gate>Use only when no specialist category fits and substantial effort spans systems/modules with broad impact. Use unspecified-low for contained moderate work.</Selection_Gate>`, DEEP_LOW_CATEGORY_CALLER_GUIDANCE = `<Selection_Gate>Route here when one subsystem plus its callers holds the mechanism and the evidence, once read, leaves one right answer. Wide but mechanical work belongs here or in a quick batch. When unsure, choose deep-low: a misrouted child returns \`ESCALATE: deep-high\` after one cheap attempt; re-spawn the same brief as deep-high with its findings.</Selection_Gate>`, DEEP_HIGH_CATEGORY_CALLER_GUIDANCE = `<Selection_Gate>Route here only when you can name the decision evidence cannot settle: a trade-off with no single right answer, a contract change crossing a package or process boundary, a mechanism with no in-repo pattern to copy, or correctness argued from invariants rather than observed in a test. Wide scope with easy decisions is deep-low or unspecified-high; reasoning as the deliverable is ultrabrain.</Selection_Gate>`, DEEP_LOW_GATE_MODELS, DEEP_HIGH_GATE_MODEL = "gpt-6-astra", OPENAI_CATEGORIES;
 var init_openai_categories = __esm(() => {
   init_types();
-  DEEP_LOW_GATE_MODELS = ["gpt-6-sol", "gpt-5.6-sol"];
+  DEEP_LOW_GATE_MODELS = ["gpt-6-sol-fast", "gpt-6-sol"];
   OPENAI_CATEGORIES = [
     {
       name: "ultrabrain",
@@ -9408,7 +9429,7 @@ var init_openai_categories = __esm(() => {
     },
     {
       name: "deep-low",
-      config: { model: "openai/gpt-6-sol", variant: "medium" },
+      config: { model: "openai/gpt-6-sol-fast", variant: "medium" },
       description: "Default deep lane: one goal, one deliverable, decisions the child can settle from what it reads. **3D graphics, computer/browser use, CAPTCHA, multimodal, backend, logic, and algorithm work is routed here.** Multiple goals fan out as parallel calls.",
       callerGuidance: DEEP_LOW_CATEGORY_CALLER_GUIDANCE,
       promptAppend: DEEP_LOW_CATEGORY_PROMPT_APPEND,
@@ -9417,7 +9438,7 @@ var init_openai_categories = __esm(() => {
     },
     {
       name: "deep-high",
-      config: { model: "openai/gpt-6-astra", variant: "high" },
+      config: { model: "openai/gpt-6-astra", variant: "xhigh" },
       description: "Escalation deep lane: a goal whose central decision cannot be settled from evidence alone. Same one-goal, one-deliverable contract as deep-low.",
       callerGuidance: DEEP_HIGH_CATEGORY_CALLER_GUIDANCE,
       promptAppend: DEEP_HIGH_CATEGORY_PROMPT_APPEND,
@@ -88811,7 +88832,7 @@ var package_default2;
 var init_package2 = __esm(() => {
   package_default2 = {
     name: "@oh-my-opencode/omo-codex",
-    version: "5.0.0-beta.85",
+    version: "5.0.0-beta.86",
     type: "module",
     private: true,
     description: "Codex harness adapter for oh-my-openagent. Vendored Codex plugin namespace (omo) + TypeScript installer + telemetry.",

package/dist/index.js

@@ -20057,26 +20057,20 @@ var AGENT_MODEL_REQUIREMENTS = {
     fallbackChain: [
       { providers: ["kimi-for-coding"], model: "kimi-for-coding-highspeed", variant: "off" },
       { providers: ["openai", "chatgpt-subscription"], model: "gpt-6-luna-fast", variant: "low" },
-      { providers: ["deepseek"], model: "deepseek-v4-flash", variant: "max" },
+      { providers: ["deepseek"], model: "deepseek-flash", variant: "max" },
       { providers: ["opencode-go", "bailian-coding-plan"], model: "qwen3.7-plus" },
-      { providers: ["opencode-go"], model: "minimax-m3" },
-      { providers: ["minimax-coding-plan", "minimax-cn-coding-plan"], model: "MiniMax-M3" },
       { providers: ["opencode-go"], model: "minimax-m2.7" },
-      { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" },
-      { providers: ["openai", "chatgpt-subscription"], model: "gpt-5.4-nano" }
+      { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" }
     ]
   },
   explore: {
     fallbackChain: [
       { providers: ["kimi-for-coding"], model: "kimi-for-coding-highspeed", variant: "off" },
       { providers: ["openai", "chatgpt-subscription"], model: "gpt-6-luna-fast", variant: "low" },
-      { providers: ["deepseek"], model: "deepseek-v4-flash", variant: "max" },
+      { providers: ["deepseek"], model: "deepseek-flash", variant: "max" },
       { providers: ["opencode-go", "bailian-coding-plan"], model: "qwen3.7-plus" },
-      { providers: ["opencode-go"], model: "minimax-m3" },
-      { providers: ["minimax-coding-plan", "minimax-cn-coding-plan"], model: "MiniMax-M3" },
       { providers: ["opencode-go"], model: "minimax-m2.7" },
-      { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" },
-      { providers: ["openai", "chatgpt-subscription"], model: "gpt-5.4-nano" }
+      { providers: ["anthropic", "github-copilot"], model: "claude-haiku-4-5" }
     ]
   },
   "multimodal-looker": {
@@ -20206,15 +20200,11 @@ var CATEGORY_MODEL_REQUIREMENTS = {
   },
   "deep-low": {
     fallbackChain: [
+      { providers: ["openai", "chatgpt-subscription"], model: "gpt-6-sol-fast", variant: "medium" },
       {
         providers: ["openai", "chatgpt-subscription", "github-copilot", "opencode"],
         model: "gpt-6-sol",
         variant: "medium"
-      },
-      {
-        providers: ["openai", "chatgpt-subscription", "github-copilot", "opencode"],
-        model: "gpt-5.6-sol",
-        variant: "medium"
       }
     ]
   },
@@ -20223,7 +20213,7 @@ var CATEGORY_MODEL_REQUIREMENTS = {
       {
         providers: ["openai", "chatgpt-subscription", "github-copilot", "opencode"],
         model: "gpt-6-astra",
-        variant: "high"
+        variant: "xhigh"
       }
     ]
   },
@@ -20249,7 +20239,7 @@ var CATEGORY_MODEL_REQUIREMENTS = {
   quick: {
     fallbackChain: [
       { providers: ["chatgpt-subscription"], model: "gpt-6-luna-fast", variant: "low" },
-      { providers: ["deepseek"], model: "deepseek-v4-flash", variant: "off" },
+      { providers: ["deepseek"], model: "deepseek-flash", variant: "off" },
       {
         providers: ["qwen-token-plan", "alibaba-token-plan", "bailian-coding-plan"],
         model: "qwen3.6-flash",
@@ -21915,6 +21905,22 @@ var SUPPLEMENTAL_MODEL_CAPABILITIES = {
       output: 128000
     }
   },
+  "gpt-6-sol-fast": {
+    id: "gpt-6-sol-fast",
+    family: "gpt",
+    reasoning: true,
+    temperature: false,
+    toolCall: true,
+    modalities: {
+      input: ["text", "image"],
+      output: ["text"]
+    },
+    limit: {
+      context: 1050000,
+      input: 922000,
+      output: 128000
+    }
+  },
   "gpt-6-luna": {
     id: "gpt-6-luna",
     family: "gpt-nano",
@@ -21995,6 +22001,21 @@ var SUPPLEMENTAL_MODEL_CAPABILITIES = {
       output: 128000
     }
   },
+  "deepseek-flash": {
+    id: "deepseek-flash",
+    family: "deepseek-flash",
+    reasoning: true,
+    temperature: true,
+    toolCall: true,
+    modalities: {
+      input: ["text", "image"],
+      output: ["text"]
+    },
+    limit: {
+      context: 1e6,
+      output: 384000
+    }
+  },
   "gpt-6-luna-fast": {
     id: "gpt-6-luna-fast",
     family: "gpt-nano",
@@ -27292,7 +27313,7 @@ You are working on tasks that don't fit specific categories but require substant
 var UNSPECIFIED_HIGH_CATEGORY_CALLER_GUIDANCE = `<Selection_Gate>Use only when no specialist category fits and substantial effort spans systems/modules with broad impact. Use unspecified-low for contained moderate work.</Selection_Gate>`;
 var DEEP_LOW_CATEGORY_CALLER_GUIDANCE = `<Selection_Gate>Route here when one subsystem plus its callers holds the mechanism and the evidence, once read, leaves one right answer. Wide but mechanical work belongs here or in a quick batch. When unsure, choose deep-low: a misrouted child returns \`ESCALATE: deep-high\` after one cheap attempt; re-spawn the same brief as deep-high with its findings.</Selection_Gate>`;
 var DEEP_HIGH_CATEGORY_CALLER_GUIDANCE = `<Selection_Gate>Route here only when you can name the decision evidence cannot settle: a trade-off with no single right answer, a contract change crossing a package or process boundary, a mechanism with no in-repo pattern to copy, or correctness argued from invariants rather than observed in a test. Wide scope with easy decisions is deep-low or unspecified-high; reasoning as the deliverable is ultrabrain.</Selection_Gate>`;
-var DEEP_LOW_GATE_MODELS = ["gpt-6-sol", "gpt-5.6-sol"];
+var DEEP_LOW_GATE_MODELS = ["gpt-6-sol-fast", "gpt-6-sol"];
 var DEEP_HIGH_GATE_MODEL = "gpt-6-astra";
 var OPENAI_CATEGORIES = [
   {
@@ -27304,7 +27325,7 @@ var OPENAI_CATEGORIES = [
   },
   {
     name: "deep-low",
-    config: { model: "openai/gpt-6-sol", variant: "medium" },
+    config: { model: "openai/gpt-6-sol-fast", variant: "medium" },
     description: "Default deep lane: one goal, one deliverable, decisions the child can settle from what it reads. **3D graphics, computer/browser use, CAPTCHA, multimodal, backend, logic, and algorithm work is routed here.** Multiple goals fan out as parallel calls.",
     callerGuidance: DEEP_LOW_CATEGORY_CALLER_GUIDANCE,
     promptAppend: DEEP_LOW_CATEGORY_PROMPT_APPEND,
@@ -27313,7 +27334,7 @@ var OPENAI_CATEGORIES = [
   },
   {
     name: "deep-high",
-    config: { model: "openai/gpt-6-astra", variant: "high" },
+    config: { model: "openai/gpt-6-astra", variant: "xhigh" },
     description: "Escalation deep lane: a goal whose central decision cannot be settled from evidence alone. Same one-goal, one-deliverable contract as deep-low.",
     callerGuidance: DEEP_HIGH_CATEGORY_CALLER_GUIDANCE,
     promptAppend: DEEP_HIGH_CATEGORY_PROMPT_APPEND,
@@ -97691,7 +97712,7 @@ function getCachedVersion(options = {}) {
 // package.json
 var package_default = {
   name: "oh-my-opencode",
-  version: "5.0.0-beta.85",
+  version: "5.0.0-beta.86",
   description: "The Best AI Agent Harness - Batteries-Included OpenCode Plugin with Multi-Model Orchestration, Parallel Background Agents, and Crafted LSP/AST Tools",
   main: "./dist/index.js",
   types: "dist/index.d.ts",
@@ -97874,7 +97895,7 @@ var package_default = {
     zod: "^4.6.5"
   },
   devDependencies: {
-    "@code-yeongyu/senpi": "2026.9.23",
+    "@code-yeongyu/senpi": "2026.9.23-2",
     "@oh-my-opencode/agents-md-core": "workspace:*",
     "@oh-my-opencode/ast-grep-mcp": "workspace:*",
     "@oh-my-opencode/boulder-state": "workspace:*",
@@ -97914,18 +97935,18 @@ var package_default = {
     typescript: "^7.0.2"
   },
   optionalDependencies: {
-    "oh-my-opencode-darwin-arm64": "5.0.0-beta.85",
-    "oh-my-opencode-darwin-x64": "5.0.0-beta.85",
-    "oh-my-opencode-darwin-x64-baseline": "5.0.0-beta.85",
-    "oh-my-opencode-linux-arm64": "5.0.0-beta.85",
-    "oh-my-opencode-linux-arm64-musl": "5.0.0-beta.85",
-    "oh-my-opencode-linux-x64": "5.0.0-beta.85",
-    "oh-my-opencode-linux-x64-baseline": "5.0.0-beta.85",
-    "oh-my-opencode-linux-x64-musl": "5.0.0-beta.85",
-    "oh-my-opencode-linux-x64-musl-baseline": "5.0.0-beta.85",
-    "oh-my-opencode-windows-arm64": "5.0.0-beta.85",
-    "oh-my-opencode-windows-x64": "5.0.0-beta.85",
-    "oh-my-opencode-windows-x64-baseline": "5.0.0-beta.85"
+    "oh-my-opencode-darwin-arm64": "5.0.0-beta.86",
+    "oh-my-opencode-darwin-x64": "5.0.0-beta.86",
+    "oh-my-opencode-darwin-x64-baseline": "5.0.0-beta.86",
+    "oh-my-opencode-linux-arm64": "5.0.0-beta.86",
+    "oh-my-opencode-linux-arm64-musl": "5.0.0-beta.86",
+    "oh-my-opencode-linux-x64": "5.0.0-beta.86",
+    "oh-my-opencode-linux-x64-baseline": "5.0.0-beta.86",
+    "oh-my-opencode-linux-x64-musl": "5.0.0-beta.86",
+    "oh-my-opencode-linux-x64-musl-baseline": "5.0.0-beta.86",
+    "oh-my-opencode-windows-arm64": "5.0.0-beta.86",
+    "oh-my-opencode-windows-x64": "5.0.0-beta.86",
+    "oh-my-opencode-windows-x64-baseline": "5.0.0-beta.86"
   },
   overrides: {
     "@earendil-works/pi-agent-core": "0.84.2",