omnilane 0.8.1 → 0.8.2

package/CHANGELOG.md

@@ -6,6 +6,19 @@ semantic version tags.
 
 ## [Unreleased]
 
+## [0.8.2] - 2026-07-18
+
+### Added
+
+- `openrouter` vendor: direct OpenRouter API dispatch (curl + `OPENROUTER_API_KEY`,
+  no CLI dependency). Advise/consult only — work mode fails hard with guidance —
+  and the model slug is mandatory. Hundreds of hosted models become reachable
+  from any omnilane install without adding a coding-agent CLI.
+- `opencode` vendor: headless `opencode run` dispatch through the OpenCode
+  multi-provider aggregator CLI. Advise mode pins OpenCode's built-in read-only
+  `plan` agent; work mode uses `--auto`. Appended as the last fallback in the
+  default `coding-overflow` chain.
+
 ## [0.8.1] - 2026-07-18
 
 ### Added
@@ -239,7 +252,8 @@ semantic version tags.
 - Initial shared routing table, cross-vendor dispatcher, runners, installer,
   and baseline lint fixes.
 
-[Unreleased]: https://github.com/Seraphim0916/omnilane/compare/v0.8.1...HEAD
+[Unreleased]: https://github.com/Seraphim0916/omnilane/compare/v0.8.2...HEAD
+[0.8.2]: https://github.com/Seraphim0916/omnilane/compare/v0.8.1...v0.8.2
 
 [0.8.1]: https://github.com/Seraphim0916/omnilane/compare/v0.8.0...v0.8.1
 

package/README.ja.md

@@ -20,6 +20,19 @@
 
 ---
 
+## v0.8.2 の新機能
+
+- **`openrouter` vendor** — `curl` と `OPENROUTER_API_KEY` だけで
+  OpenRouter API に直接ディスパッチ。どの omnilane インストールからでも
+  数百のホスト型モデルに到達でき、コーディングエージェント CLI の追加は
+  不要。advise/consult 専用(ファイル編集不可、work モードは明確に
+  エラーで案内)、モデル slug は必須。例:
+  `dispatch.sh --vendor openrouter --model anthropic/claude-sonnet-5 consult "..."`
+- **`opencode` vendor** — マルチプロバイダ集約 CLI OpenCode 経由の
+  ヘッドレスディスパッチ(`opencode run`)。advise モードは組み込みの
+  読み取り専用 `plan` エージェントを使用し、work モードは `--auto`。
+  デフォルトの `coding-overflow` チェーンの最終フォールバックに追加。
+
 ## v0.8.1 の新機能
 
 - **Claude Code プラグインがルーティングリマインダーを自動読み込み** —
@@ -149,7 +162,7 @@ flowchart LR
 | 📚 long-context | Gemini 3.1 Pro (High) | Claude Opus 4.8 (high) | 100 万トークン級の長文統合——分析専用、agentic ループ禁止 |
 | ⚡ fast-agentic | Gemini 3.5 Flash (High) | GPT-5.6 Luna (high) | 高速なマルチステップ agentic ループ、マルチモーダル確認 |
 | 📡 live-search | Grok 4.5 | —(off) | リアルタイム X/ウェブ検索とソーシャル文脈 |
-| 🚰 coding-overflow | Grok 4.5 | Kimi K3 → Qwen3 Coder Plus | Codex クォータ逼迫時の中級コーディング逃し弁 |
+| 🚰 coding-overflow | Grok 4.5 | Kimi K3 → Qwen3 Coder Plus → OpenCode | Codex クォータ逼迫時の中級コーディング逃し弁 |
 | 🗳️ arbitrate | off(オプトイン) | — | 内蔵オピニオンパネル(重大な判断用)——デフォルト無効。`routing.local.yaml` で有効化;投票者×ラウンドごとに 1 コール消費 |
 
 **バックアップ**はチェーンの次の候補——第一候補のベンダー CLI が未インストールの

package/README.ko.md

@@ -20,6 +20,19 @@
 
 ---
 
+## v0.8.2 새 기능
+
+- **`openrouter` vendor** — `curl`과 `OPENROUTER_API_KEY`만으로
+  OpenRouter API에 직접 디스패치합니다. 어떤 omnilane 설치에서도 수백 개의
+  호스팅 모델에 접근할 수 있으며 코딩 에이전트 CLI를 추가로 설치할 필요가
+  없습니다. advise/consult 전용(파일 편집 불가, work 모드는 명확한 오류로
+  안내)이며 모델 slug는 필수입니다. 예:
+  `dispatch.sh --vendor openrouter --model anthropic/claude-sonnet-5 consult "..."`
+- **`opencode` vendor** — 멀티 프로바이더 집합 CLI OpenCode를 통한
+  헤드리스 디스패치(`opencode run`). advise 모드는 내장 읽기 전용 `plan`
+  에이전트를 사용하고 work 모드는 `--auto`를 사용합니다. 기본
+  `coding-overflow` 체인의 마지막 폴백으로 추가되었습니다.
+
 ## v0.8.1 새 기능
 
 - **Claude Code 플러그인이 라우팅 리마인더를 자동 로드** — 플러그인에
@@ -148,7 +161,7 @@ flowchart LR
 | 📚 long-context | Gemini 3.1 Pro (High) | Claude Opus 4.8 (high) | 100만 토큰급 장문 통합——분석 전용, agentic 루프 금지 |
 | ⚡ fast-agentic | Gemini 3.5 Flash (High) | GPT-5.6 Luna (high) | 빠른 멀티스텝 agentic 루프, 멀티모달 확인 |
 | 📡 live-search | Grok 4.5 | —(off) | 실시간 X/웹 검색과 소셜 맥락 |
-| 🚰 coding-overflow | Grok 4.5 | Kimi K3 → Qwen3 Coder Plus | Codex 쿼터 소진 시 중급 코딩 안전 밸브 |
+| 🚰 coding-overflow | Grok 4.5 | Kimi K3 → Qwen3 Coder Plus → OpenCode | Codex 쿼터 소진 시 중급 코딩 안전 밸브 |
 | 🗳️ arbitrate | off(옵트인) | — | 내장 의견 패널(중대한 결정용)——기본 비활성. `routing.local.yaml` 에서 활성화;투표자×라운드마다 1콜 소모 |
 
 **백업**은 체인의 다음 후보입니다——1순위 벤더 CLI 가 설치되지 않았을 때

package/README.md

@@ -20,6 +20,19 @@ Every subtask goes to the model that is actually best at it — across<br/>
 
 ---
 
+## What's new in v0.8.2
+
+- **`openrouter` vendor** — dispatch straight to the OpenRouter API with
+  nothing but `curl` and an `OPENROUTER_API_KEY`: hundreds of hosted models
+  become reachable from any omnilane install, no coding-agent CLI required.
+  Advise/consult only (it cannot edit files; work mode fails with guidance)
+  and the model slug is mandatory, e.g.
+  `dispatch.sh --vendor openrouter --model anthropic/claude-sonnet-5 consult "..."`.
+- **`opencode` vendor** — headless dispatch through the OpenCode
+  multi-provider aggregator CLI (`opencode run`). Advise mode pins OpenCode's
+  built-in read-only `plan` agent; work mode uses `--auto`. Joins the default
+  `coding-overflow` chain as its last fallback.
+
 ## What's new in v0.8.1
 
 - **Claude Code plugin auto-loads the routing reminder** — the plugin now
@@ -150,7 +163,7 @@ flowchart LR
 | 📚 long-context | Gemini 3.1 Pro (High) | Claude Opus 4.8 (high) | 1M-token synthesis — analysis only, never agentic loops |
 | ⚡ fast-agentic | Gemini 3.5 Flash (High) | GPT-5.6 Luna (high) | Fast multi-step agentic loops, multimodal checks |
 | 📡 live-search | Grok 4.5 | — (off) | Realtime X/web search and social context |
-| 🚰 coding-overflow | Grok 4.5 | Kimi K3 → Qwen3 Coder Plus | Codex-quota relief valve for mid-tier coding |
+| 🚰 coding-overflow | Grok 4.5 | Kimi K3 → Qwen3 Coder Plus → OpenCode | Codex-quota relief valve for mid-tier coding |
 | 🗳️ arbitrate | off (opt-in vote panel) | — | Built-in opinion panel for big calls — disabled by default; enable it in `routing.local.yaml`, one call per voter per round |
 
 The **backup** is the next candidate in the lane's `routing.yaml` chain — what

package/README.zh-CN.md

@@ -20,6 +20,17 @@
 
 ---
 
+## v0.8.2 新功能
+
+- **`openrouter` vendor** — 只需 `curl` 加一个 `OPENROUTER_API_KEY`,
+  即可直连 OpenRouter API 派工:任何 omnilane 安装都能访问数百个
+  托管模型,无需再装任何代理 CLI。仅限 advise/consult(不能改文件,
+  work 模式会明确报错指路),模型 slug 必填,例如
+  `dispatch.sh --vendor openrouter --model anthropic/claude-sonnet-5 consult "..."`。
+- **`opencode` vendor** — 通过 OpenCode 多供应商聚合 CLI 无头派工
+  (`opencode run`)。advise 模式锁定内置只读 `plan` agent;work 模式
+  用 `--auto`。加入默认 `coding-overflow` 链作为最后回退。
+
 ## v0.8.1 新功能
 
 - **Claude Code 插件开场自动载入路由提醒** — 插件新增 `SessionStart`
@@ -134,7 +145,7 @@ flowchart LR
 | 📚 long-context | Gemini 3.1 Pro (High) | Claude Opus 4.8 (high) | 百万 token 长文整合——仅限分析,不派 agentic 长链 |
 | ⚡ fast-agentic | Gemini 3.5 Flash (High) | GPT-5.6 Luna (high) | 快速多步骤 agentic 循环、多模态检查 |
 | 📡 live-search | Grok 4.5 | —(off) | 实时 X/网络搜索与社群脉络 |
-| 🚰 coding-overflow | Grok 4.5 | Kimi K3 → Qwen3 Coder Plus | Codex 额度吃紧时的中量级编码溢流道;事实性声明须另行查证 |
+| 🚰 coding-overflow | Grok 4.5 | Kimi K3 → Qwen3 Coder Plus → OpenCode | Codex 额度吃紧时的中量级编码溢流道;事实性声明须另行查证 |
 | 🗳️ arbitrate | off(可选评审团) | — | 内置意见评审团,重大决定用——默认关闭,要用在 `routing.local.yaml` 打开;每评审每轮烧一次额度 |
 
 **备选模型**是候选链的下一位——首选那家的厂商 CLI 没装时,派发就降到它。

package/README.zh-TW.md

@@ -20,6 +20,17 @@
 
 ---
 
+## v0.8.2 新功能
+
+- **`openrouter` vendor** — 只要 `curl` 加一把 `OPENROUTER_API_KEY`,
+  就能直連 OpenRouter API 派工:任何 omnilane 安裝都摸得到數百個
+  託管模型,不必再裝任何代理 CLI。僅限 advise/consult(不能改檔,
+  work 模式會明確報錯指路),模型 slug 必填,例如
+  `dispatch.sh --vendor openrouter --model anthropic/claude-sonnet-5 consult "..."`。
+- **`opencode` vendor** — 透過 OpenCode 多供應商聚合 CLI 無頭派工
+  (`opencode run`)。advise 模式鎖定內建唯讀 `plan` agent;work 模式
+  用 `--auto`。加入預設 `coding-overflow` 鏈作為最後備援。
+
 ## v0.8.1 新功能
 
 - **Claude Code 外掛開場自動載入路由提醒** — 外掛新增 `SessionStart`
@@ -134,7 +145,7 @@ flowchart LR
 | 📚 long-context | Gemini 3.1 Pro (High) | Claude Opus 4.8 (high) | 百萬 token 長文整合——僅限分析,不派 agentic 長鏈 |
 | ⚡ fast-agentic | Gemini 3.5 Flash (High) | GPT-5.6 Luna (high) | 快速多步驟 agentic 迴圈、多模態檢查 |
 | 📡 live-search | Grok 4.5 | —(off) | 即時 X/網路搜尋與社群脈絡 |
-| 🚰 coding-overflow | Grok 4.5 | Kimi K3 → Qwen3 Coder Plus | Codex 額度吃緊時的中量級編碼溢流道;事實性宣稱須另行查證 |
+| 🚰 coding-overflow | Grok 4.5 | Kimi K3 → Qwen3 Coder Plus → OpenCode | Codex 額度吃緊時的中量級編碼溢流道;事實性宣稱須另行查證 |
 | 🗳️ arbitrate | off(選配評審團) | — | 內建意見評審團,重大決定用——預設關閉,要用在 `routing.local.yaml` 開;每評審每輪燒一次額度 |
 
 **備選模型**是候選鏈的下一位——首選那家的廠商 CLI 沒裝時,派工就降到它。

package/VERSION

@@ -1 +1 @@
-0.8.1
+0.8.2

package/completions/_omnilane

@@ -64,7 +64,7 @@ _omnilane() {
         '--dry-run[resolve and print the plan without side effects]' \
         '--mode[execution mode]:mode:(advise work)' \
         '--workdir[target directory]:directory:_directories' \
-        '--vendor[vendor override]:vendor:(codex claude grok gemini exec vote)' \
+        '--vendor[vendor override]:vendor:(codex claude grok gemini kimi qwen opencode openrouter exec vote)' \
         '--model[model override]:model:' \
         '--effort[reasoning effort]:effort:(low medium high xhigh max)' \
         '--timeout[per-call seconds]:seconds:' \

package/completions/omnilane.bash

@@ -64,7 +64,7 @@ _omnilane() {
       route|dispatch)
         case "$prev" in
           --mode) words="advise work" ;;
-          --vendor) words="codex claude grok gemini kimi qwen exec vote" ;;
+          --vendor) words="codex claude grok gemini kimi qwen opencode openrouter exec vote" ;;
           --effort) words="low medium high xhigh max" ;;
           --workdir)
             COMPREPLY=()

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "omnilane",
-  "version": "0.8.1",
+  "version": "0.8.2",
   "description": "One routing table, every harness — classify subtasks into lanes and dispatch each lane to the best vendor's agentic CLI (Codex, Claude, Gemini, Grok) using your existing subscription logins.",
   "bin": {
     "omnilane": "bin/omnilane"

package/routing.yaml

@@ -1,6 +1,8 @@
 # omnilane — single routing table shared by every harness (CC / Codex / Grok / Antigravity)
 # Format: one lane per line ->  lane: vendor model effort [| vendor model effort | off]
-#   vendor  = codex | claude | grok | gemini
+#   vendor  = codex | claude | grok | gemini | kimi | qwen | opencode | openrouter
+#   openrouter is direct-API (needs only OPENROUTER_API_KEY + curl) and advise/consult
+#   only — it cannot edit files, so it is not in the default work-capable chains.
 #   effort  = codex reasoning effort / claude --effort / agy thinking suffix / (grok: ignored)
 #   Fallback chain: the first candidate whose vendor CLI is installed wins, so the
 #   same table degrades gracefully when you only subscribe to one or two vendors.
@@ -22,7 +24,7 @@ ui-draft:        codex gpt-5.6-sol xhigh | claude claude-opus-4-8 high  # only w
 long-context:    gemini "Gemini 3.1 Pro (High)" - | claude claude-opus-4-8 high | codex gpt-5.6-sol high  # 1M-token synthesis ONLY — never agentic loops (AA agentic 21.4)
 fast-agentic:    gemini "Gemini 3.5 Flash (High)" - | codex gpt-5.6-luna high  # speed + agentic tool loops
 live-search:     grok grok-4.5 - | off          # native X/web search lane; no real substitute
-coding-overflow: grok grok-4.5 - | kimi kimi-k3 - | qwen qwen3-coder-plus - | off  # codex-quota relief valve: mid-tier coding; verify factual claims (grok hallucination-prone). kimi/qwen model fields are CLI aliases — adjust to your login.
+coding-overflow: grok grok-4.5 - | kimi kimi-k3 - | qwen qwen3-coder-plus - | opencode - - | off  # codex-quota relief valve: mid-tier coding; verify factual claims (grok hallucination-prone). kimi/qwen model fields are CLI aliases — adjust to your login. opencode "-" model = its own configured default.
 arbitrate:       off - -                        # opinion panel is OPT-IN: it costs one call per voter per round.
                                                 # Enable: `arbitrate: vote codex,claude,grok -` (any 1-4 of codex/claude/grok/gemini)
                                                 # Debate round (each voter rebuts the others): set the effort field to 2.

package/scripts/configure.sh

@@ -18,6 +18,9 @@ GEMINI_MODELS=("Gemini 3.1 Pro (High)" "Gemini 3.1 Pro (Low)" "Gemini 3.5 Flash
 GROK_MODELS=("grok-4.5" "grok-4.3")
 KIMI_MODELS=("kimi-k3" "kimi-k2.7-code")
 QWEN_MODELS=("qwen3-coder-plus" "qwen3-coder-flash")
+# OpenCode models use provider/model form; OpenRouter models are catalog slugs.
+OPENCODE_MODELS=("openrouter/anthropic/claude-sonnet-5" "openrouter/openai/gpt-5.6-sol" "opencode/default (leave model to opencode)")
+OPENROUTER_MODELS=("anthropic/claude-sonnet-5" "openai/gpt-5.6-sol" "moonshotai/kimi-k3" "qwen/qwen3-coder-plus")
 
 custom_value_is_safe() {
   case "$1" in
@@ -76,7 +79,7 @@ while true; do
     echo "$(msgf cfg_pick_range "${#LANES[@]}")"; continue
   fi
   lane="${LANES[$((n - 1))]}"
-  vendor="$(pick "$(msgf cfg_vendor_for "$lane")" codex claude grok gemini kimi qwen "vote (multi-model panel)" "exec (your own script/gate)" off)"
+  vendor="$(pick "$(msgf cfg_vendor_for "$lane")" codex claude grok gemini kimi qwen opencode openrouter "vote (multi-model panel)" "exec (your own script/gate)" off)"
   [[ "$vendor" == exec* ]] && vendor="exec"
   [[ "$vendor" == vote* ]] && vendor="vote"
   if [[ "$vendor" == "off" ]]; then
@@ -90,6 +93,9 @@ while true; do
     grok)   model="$(pick "$(msg cfg_model)" "${GROK_MODELS[@]}")";   effort="-" ;;
     kimi)   model="$(pick "$(msg cfg_model)" "${KIMI_MODELS[@]}")";   effort="-" ;;
     qwen)   model="$(pick "$(msg cfg_model)" "${QWEN_MODELS[@]}")";   effort="-" ;;
+    opencode)   model="$(pick "$(msg cfg_model)" "${OPENCODE_MODELS[@]}")"; effort="-"
+                [[ "$model" == "opencode/default"* ]] && model="-" ;;
+    openrouter) model="$(pick "$(msg cfg_model)" "${OPENROUTER_MODELS[@]}")"; effort="-" ;;
     vote)   while true; do
               count="$(pick "$(msg cfg_voters_count)" "1" "2" "3" "4")"
               [[ "$count" =~ ^[1-4]$ ]] && break

package/scripts/dispatch.sh

@@ -52,7 +52,7 @@ flags:
                          before any provider call or job state
   --mode advise|work     advise (read-only, default) or work (may edit files)
   --workdir DIR          working directory handed to the vendor CLI
-  --vendor V             pin one configured vendor (codex|claude|grok|gemini|kimi|qwen)
+  --vendor V             pin one configured vendor (codex|claude|grok|gemini|kimi|qwen|opencode|openrouter)
   --model M              override the routed model
   --effort E             override the routed effort
   --timeout SECONDS      cap each CLI call (default 600)
@@ -363,7 +363,7 @@ validate_routing() {
           lane_invalid=1
           break
         fi
-        if ! [[ "$vendor" =~ ^(codex|claude|grok|gemini|kimi|qwen|exec|off)$ ]]; then
+        if ! [[ "$vendor" =~ ^(codex|claude|grok|gemini|kimi|qwen|opencode|openrouter|exec|off)$ ]]; then
           printf 'FAIL %s unknown-vendor=%s\n' "$lane" "$vendor"
           lane_invalid=1
           break
@@ -489,8 +489,8 @@ TASK="$2"
 # A typo like --mode advice must not fall through to the write-enabled branch.
 [[ "$MODE" == "advise" || "$MODE" == "work" ]] || { echo "omnilane: invalid --mode (advise|work)" >&2; exit 2; }
 if [[ -n "$OVERRIDE_VENDOR" ]] &&
-   ! [[ "$OVERRIDE_VENDOR" =~ ^(codex|claude|grok|gemini|kimi|qwen)$ ]]; then
-  echo "omnilane: invalid vendor (codex|claude|grok|gemini|kimi|qwen)" >&2
+   ! [[ "$OVERRIDE_VENDOR" =~ ^(codex|claude|grok|gemini|kimi|qwen|opencode|openrouter)$ ]]; then
+  echo "omnilane: invalid vendor (codex|claude|grok|gemini|kimi|qwen|opencode|openrouter)" >&2
   exit 2
 fi
 

package/scripts/jobs.sh

@@ -132,7 +132,7 @@ read_job_pid() {
 
 parse_stats_metadata() {
   local value="$1"
-  local metadata_re='^\{"lane":"([a-z][a-z0-9-]*)","vendor":"(codex|claude|grok|gemini|kimi|qwen|exec)"(,|})'
+  local metadata_re='^\{"lane":"([a-z][a-z0-9-]*)","vendor":"(codex|claude|grok|gemini|kimi|qwen|opencode|openrouter|exec)"(,|})'
   STATS_LANE=""
   STATS_VENDOR=""
   [[ "$value" =~ $metadata_re ]] || return 1
@@ -143,7 +143,7 @@ parse_stats_metadata() {
 parse_audit_metadata() {
   local value="$1" metadata_re
   local json_string='([^"\\]|\\["\\/bfnrt]|\\u[0-9a-fA-F]{4})*'
-  metadata_re="^\\{\"lane\":\"[a-z][a-z0-9-]*\",\"vendor\":\"(codex|claude|grok|gemini|kimi|qwen|exec)\",\"model\":\"${json_string}\",\"effort\":\"${json_string}\",\"timeout\":(0|[1-9][0-9]*),\"job_timeout\":(null|0|[1-9][0-9]*),\"mode\":\"(advise|work)\",\"workdir\":\"${json_string}\",\"candidate\":\"[1-9][0-9]*/[1-9][0-9]*\",\"started\":\"[0-9]{4}-[0-9]{2}-[0-9]{2}T[0-9]{2}:[0-9]{2}:[0-9]{2}Z\"\\}$"
+  metadata_re="^\\{\"lane\":\"[a-z][a-z0-9-]*\",\"vendor\":\"(codex|claude|grok|gemini|kimi|qwen|opencode|openrouter|exec)\",\"model\":\"${json_string}\",\"effort\":\"${json_string}\",\"timeout\":(0|[1-9][0-9]*),\"job_timeout\":(null|0|[1-9][0-9]*),\"mode\":\"(advise|work)\",\"workdir\":\"${json_string}\",\"candidate\":\"[1-9][0-9]*/[1-9][0-9]*\",\"started\":\"[0-9]{4}-[0-9]{2}-[0-9]{2}T[0-9]{2}:[0-9]{2}:[0-9]{2}Z\"\\}$"
   [[ "$value" =~ $metadata_re ]]
 }
 
@@ -252,7 +252,7 @@ case "${1:-}" in
     read_public_metadata "$JOB_DIR/meta.json" || {
       die 1 "cannot retry: unreadable job metadata"
     }
-    retry_re='^\{"lane":"([a-z][a-z0-9-]*)","vendor":"(codex|claude|grok|gemini|kimi|qwen|exec)","model":"([^"\\]*)","effort":"([^"\\]*)","timeout":([1-9][0-9]*),"job_timeout":([1-9][0-9]*|null),"mode":"(advise|work)","workdir":"([^"\\]*)",'
+    retry_re='^\{"lane":"([a-z][a-z0-9-]*)","vendor":"(codex|claude|grok|gemini|kimi|qwen|opencode|openrouter|exec)","model":"([^"\\]*)","effort":"([^"\\]*)","timeout":([1-9][0-9]*),"job_timeout":([1-9][0-9]*|null),"mode":"(advise|work)","workdir":"([^"\\]*)",'
     [[ "$PUBLIC_METADATA" =~ $retry_re ]] || {
       die 1 "cannot retry: job metadata is not safely parseable"
     }

package/scripts/lib/common.sh

@@ -72,6 +72,7 @@ vendor_bin() { # vendor -> binary (honors local.sh overrides)
     gemini) echo "${AGY_BIN:-agy}" ;;
     kimi)   echo "${KIMI_BIN:-kimi}" ;;
     qwen)   echo "${QWEN_BIN:-qwen}" ;;
+    opencode) echo "${OPENCODE_BIN:-opencode}" ;;
     *)      echo "" ;;
   esac
 }
@@ -81,6 +82,11 @@ vendor_available() {
   # checked at dispatch time (the chain resolver cannot see it here).
   # "vote" is the built-in multi-model panel: needs >=2 voters, checked at dispatch.
   [[ "$1" == "exec" || "$1" == "vote" ]] && return 0
+  # "openrouter" is direct-API (no CLI): available iff curl and a key exist.
+  if [[ "$1" == "openrouter" ]]; then
+    command -v curl >/dev/null 2>&1 && [[ -n "${OPENROUTER_API_KEY:-}" ]]
+    return
+  fi
   local b; b="$(vendor_bin "$1")"
   [[ -n "$b" ]] && command -v "$b" >/dev/null 2>&1
 }

package/scripts/runners/run-opencode.sh

@@ -0,0 +1,51 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# omnilane runner: OpenCode CLI (multi-provider aggregator)
+#
+# Usage: run-opencode.sh MODE WORKDIR MODEL EFFORT PROMPT_FILE OUTPUT_FILE
+#   MODE = advise (built-in read-only plan agent) | work (may edit files)
+#
+# MODEL is OpenCode's provider/model form (e.g. openrouter/moonshotai/kimi-k2);
+# "-" lets OpenCode use its own configured default. OpenCode's --variant flag
+# is provider-specific, so EFFORT maps onto it only when set.
+
+source "$(dirname "${BASH_SOURCE[0]}")/../lib/common.sh"
+
+MODE="$1"; WORKDIR="$2"; MODEL="$3"; EFFORT="$4"; PROMPT_FILE="$5"; OUTPUT_FILE="$6"
+
+OPENCODE_BIN="${OPENCODE_BIN:-opencode}"
+RUN_TIMEOUT="${OMNILANE_TIMEOUT:-600}"
+
+truncate_payload "$PROMPT_FILE" 140000
+
+# --agent plan is OpenCode's built-in restricted agent (edits/bash gated);
+# --auto only in work mode, and only for permissions not explicitly denied.
+ARGS=(run "$(cat "$PROMPT_FILE")" --format default)
+[[ -n "$MODEL" && "$MODEL" != "-" ]] && ARGS+=(-m "$MODEL")
+[[ -n "$EFFORT" && "$EFFORT" != "-" ]] && ARGS+=(--variant "$EFFORT")
+if [[ "$MODE" == "advise" ]]; then
+  ARGS+=(--agent plan)
+else
+  ARGS+=(--auto)
+fi
+
+set +e
+(
+  cd "$WORKDIR" || exit 127
+  OMNILANE_DEPTH=1 run_with_timeout "$RUN_TIMEOUT" \
+    "$OPENCODE_BIN" "${ARGS[@]}" \
+    > "${OUTPUT_FILE}.tmp" 2> "${OUTPUT_FILE}.stderr.log"
+)
+RC=$?
+set -e
+
+# Empty output is a failure, not a silent rc=0 success.
+if ! grep -q '[^[:space:]]' "${OUTPUT_FILE}.tmp" 2>/dev/null; then
+  echo "omnilane: opencode produced no output" >> "${OUTPUT_FILE}.stderr.log"
+  [[ "$RC" -eq 0 ]] && RC=1
+fi
+
+[[ -f "${OUTPUT_FILE}.tmp" ]] && mv "${OUTPUT_FILE}.tmp" "$OUTPUT_FILE"
+[[ -s "${OUTPUT_FILE}.stderr.log" ]] || rm "${OUTPUT_FILE}.stderr.log" 2>/dev/null || true
+exit "$RC"

package/scripts/runners/run-openrouter.sh

@@ -0,0 +1,86 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# omnilane runner: OpenRouter direct API (no CLI dependency)
+#
+# Usage: run-openrouter.sh MODE WORKDIR MODEL EFFORT PROMPT_FILE OUTPUT_FILE
+#   MODE = advise only. This vendor is pure inference over HTTPS — it has no
+#   agentic loop and cannot edit files, so work mode is a hard error rather
+#   than a silent read-only downgrade.
+#
+# MODEL is an OpenRouter model slug (e.g. anthropic/claude-sonnet-4). EFFORT
+# is accepted for interface parity only. Needs OPENROUTER_API_KEY and curl;
+# python3 builds/parses the JSON so prompt content never needs shell escaping.
+
+source "$(dirname "${BASH_SOURCE[0]}")/../lib/common.sh"
+
+MODE="$1"; WORKDIR="$2"; MODEL="$3"; EFFORT="$4"; PROMPT_FILE="$5"; OUTPUT_FILE="$6"
+: "$WORKDIR" "$EFFORT" # parity with the uniform runner interface
+
+OPENROUTER_BASE_URL="${OPENROUTER_BASE_URL:-https://openrouter.ai/api/v1}"
+RUN_TIMEOUT="${OMNILANE_TIMEOUT:-600}"
+
+fail() { echo "omnilane: $1" > "${OUTPUT_FILE}.stderr.log"; exit 2; }
+
+[[ "$MODE" == "advise" ]] || fail "openrouter is inference-only: use --mode advise (work mode needs an agentic CLI vendor like opencode)"
+[[ -n "${OPENROUTER_API_KEY:-}" ]] || fail "OPENROUTER_API_KEY is not set"
+[[ -n "$MODEL" && "$MODEL" != "-" ]] || fail "openrouter needs an explicit model slug (e.g. anthropic/claude-sonnet-4)"
+command -v curl >/dev/null 2>&1 || fail "curl not found"
+command -v python3 >/dev/null 2>&1 || fail "python3 not found"
+
+truncate_payload "$PROMPT_FILE" 140000
+
+REQUEST_FILE="${OUTPUT_FILE}.request.json"
+RESPONSE_FILE="${OUTPUT_FILE}.response.json"
+cleanup() { rm "$REQUEST_FILE" "$RESPONSE_FILE" 2>/dev/null || true; }
+trap cleanup EXIT
+
+python3 - "$MODEL" "$PROMPT_FILE" > "$REQUEST_FILE" <<'PY'
+import json, sys
+model, prompt_file = sys.argv[1], sys.argv[2]
+with open(prompt_file, encoding="utf-8", errors="replace") as f:
+    prompt = f.read()
+json.dump({"model": model, "messages": [{"role": "user", "content": prompt}]},
+          sys.stdout)
+PY
+
+set +e
+run_with_timeout "$RUN_TIMEOUT" curl -sS --fail-with-body \
+  --max-time "$RUN_TIMEOUT" \
+  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
+  -H "Content-Type: application/json" \
+  -X POST "$OPENROUTER_BASE_URL/chat/completions" \
+  --data-binary "@$REQUEST_FILE" \
+  > "$RESPONSE_FILE" 2> "${OUTPUT_FILE}.stderr.log"
+RC=$?
+set -e
+
+if [[ "$RC" -ne 0 ]]; then
+  # Surface the API error body (never the key) alongside curl's own stderr.
+  [[ -s "$RESPONSE_FILE" ]] && cat "$RESPONSE_FILE" >> "${OUTPUT_FILE}.stderr.log"
+  exit "$RC"
+fi
+
+set +e
+python3 - "$RESPONSE_FILE" > "${OUTPUT_FILE}.tmp" 2>> "${OUTPUT_FILE}.stderr.log" <<'PY'
+import json, sys
+with open(sys.argv[1], encoding="utf-8", errors="replace") as f:
+    data = json.load(f)
+if "error" in data:
+    print(f"omnilane: openrouter API error: {data['error']}", file=sys.stderr)
+    sys.exit(1)
+content = data["choices"][0]["message"]["content"]
+sys.stdout.write(content if isinstance(content, str) else json.dumps(content))
+PY
+RC=$?
+set -e
+
+# Empty output is a failure, not a silent rc=0 success.
+if ! grep -q '[^[:space:]]' "${OUTPUT_FILE}.tmp" 2>/dev/null; then
+  echo "omnilane: openrouter produced no output" >> "${OUTPUT_FILE}.stderr.log"
+  [[ "$RC" -eq 0 ]] && RC=1
+fi
+
+[[ -f "${OUTPUT_FILE}.tmp" ]] && mv "${OUTPUT_FILE}.tmp" "$OUTPUT_FILE"
+[[ -s "${OUTPUT_FILE}.stderr.log" ]] || rm "${OUTPUT_FILE}.stderr.log" 2>/dev/null || true
+exit "$RC"