fix: replace ollama launch claude with direct env var approach

ollama launch claude requires a TTY, which breaks in non-interactive
contexts (background processes, CI). Instead, point Claude Code
directly at Ollama's Anthropic-compatible API using env vars:
ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_BASE_URL=https://ollama.com

Author: caiopizzol

README.md

@@ -127,7 +127,7 @@ Review results are saved as JSON files containing raw model outputs, timestamps,
 | Grok    | `grok-code-fast-1`, `grok-4-1-fast-*`, `grok-4-fast-*`, `grok-3`, `grok-3-mini`                                                 | [xAI API Models](https://docs.x.ai/docs/models)                |
 | Ollama  | Cloud (recommended): `minimax-m2.5:cloud`, `glm-5:cloud`, `kimi-k2.5:cloud`, or any model from library | [Ollama Library](https://ollama.com/library) |
 
-> **Note:** Ollama cloud models (`:cloud` suffix) run via `ollama launch claude` with full agentic capabilities — tools, web search, and subagents. Requires `OLLAMA_API_KEY`. Get your API key at [ollama.com](https://ollama.com). Local models (e.g., `qwen2.5-coder:7b`) use `ollama run` (text-only), are slower, and require significant memory (~8GB+ RAM for 7B models).
+> **Note:** Ollama cloud models (`:cloud` suffix) run Claude Code pointed at Ollama's Anthropic-compatible API (`ANTHROPIC_BASE_URL=https://ollama.com`). This gives cloud models full agentic capabilities — tools, file access, and subagents. Requires `OLLAMA_API_KEY`. Get your API key at [ollama.com](https://ollama.com). Local models (e.g., `qwen2.5-coder:7b`) use `ollama run` (text-only), are slower, and require significant memory (~8GB+ RAM for 7B models).
 
 > **Note:** Mistral and Grok use command-line argument passing (not stdin), which has a ~200KB limit on macOS. Very large diffs may cause these tools to fail while other tools succeed.
 
@@ -152,7 +152,7 @@ Customize prompts for each command:
 | Qwen    | `npm install -g @qwen-code/qwen-code`                                         |
 | Mistral | `pipx install mistral-vibe`                                                   |
 | Grok    | `bun add -g @vibe-kit/grok-cli`; `export GROK_API_KEY="key"` in `~/.zshrc`    |
-| Ollama  | [ollama.com/download](https://ollama.com/download); cloud (agentic): `export OLLAMA_API_KEY="key"` in `~/.zshrc`; local: `ollama pull <model>` |
+| Ollama  | [ollama.com/download](https://ollama.com/download); cloud: `export OLLAMA_API_KEY="key"` in `~/.zshrc`; local: `ollama pull <model>` |
 
 ## Usage
 

agents/multi-model-executor.md

@@ -43,14 +43,14 @@ For each tool in the config:
 
 **Environment override for nested Claude Code**:
 
-When running inside Claude Code, `CLAUDECODE=1` prevents spawning nested sessions. Prefix with `CLAUDECODE=0` for any tool whose command starts with `claude` or uses `ollama launch claude`:
+When running inside Claude Code, `CLAUDECODE=1` prevents spawning nested sessions. Prefix with `CLAUDECODE=0` for any tool whose command contains `claude`:
 
 ```bash
 # Claude tools
 cat /tmp/conclave-prompt.md | CLAUDECODE=0 claude --print 2>&1
 
-# Ollama cloud tools (ollama launch claude wraps Claude Code)
-cat /tmp/conclave-prompt.md | CLAUDECODE=0 ollama launch claude --model minimax-m2.5:cloud -- --print 2>&1
+# Ollama cloud tools (runs Claude Code pointed at Ollama's API)
+cat /tmp/conclave-prompt.md | CLAUDECODE=0 ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print --model glm-5:cloud 2>&1
 ```
 
 **Stdin-based tools** (most):
@@ -83,19 +83,18 @@ Skip injection if command already contains a model flag.
 
 Ollama has two command patterns depending on model type:
 
-**Cloud models** (`:cloud` suffix) — use `ollama launch claude` (agentic, with tools/web search):
-- Command: `ollama launch claude -- --print`
-- Model injection: `--model` flag inserted before `--`
-- Requires `CLAUDECODE=0` prefix
-- Example: `cat /tmp/conclave-prompt.md | CLAUDECODE=0 ollama launch claude --model qwen3-coder:480b-cloud -- --print 2>&1`
+**Cloud models** (`:cloud` suffix) — run Claude Code pointed at Ollama's Anthropic-compatible API:
+- Command: `ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print`
+- Model injection: `--model` flag appended (same as regular Claude)
+- Requires `CLAUDECODE=0` prefix (since it runs `claude`)
+- Requires `OLLAMA_API_KEY` env var
+- Example: `cat /tmp/conclave-prompt.md | CLAUDECODE=0 ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print --model glm-5:cloud 2>&1`
 
 **Local models** (no `:cloud` suffix) — use `ollama run` (text-only):
 - Command: `ollama run`
 - Model injection: Appended directly (no flag)
 - Example: `cat /tmp/conclave-prompt.md | ollama run qwen2.5-coder:7b 2>&1`
 
-Detection: If the tool's `model` field ends with `:cloud`, use the cloud pattern.
-
 ### Step 3: Collect Results
 
 Use TaskOutput for each background task to wait for completion.

commands/consult.md

@@ -125,14 +125,14 @@ PROMPT_EOF
 
 **Step 5b - Run consultation commands in background** (run ALL in parallel with `run_in_background: true`):
 
-**Environment override for nested Claude Code**: When running inside Claude Code, `CLAUDECODE=1` prevents spawning nested sessions. Prefix with `CLAUDECODE=0` for any tool whose command starts with `claude` or uses `ollama launch claude`:
+**Environment override for nested Claude Code**: When running inside Claude Code, `CLAUDECODE=1` prevents spawning nested sessions. Prefix with `CLAUDECODE=0` for any tool whose command contains `claude`:
 
 ```bash
 # Claude tools
 cat /tmp/conclave-consult-prompt.md | CLAUDECODE=0 claude --print --model opus 2>&1
 
-# Ollama cloud tools
-cat /tmp/conclave-consult-prompt.md | CLAUDECODE=0 ollama launch claude --model minimax-m2.5:cloud -- --print 2>&1
+# Ollama cloud tools (runs Claude Code pointed at Ollama's API)
+cat /tmp/conclave-consult-prompt.md | CLAUDECODE=0 ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print --model glm-5:cloud 2>&1
 ```
 
 For stdin-based tools:
@@ -244,7 +244,7 @@ Same as `/review` - see `~/.config/conclave/tools.json` for enabled tools.
 | Mistral  | `vibe --output text -p`            | Config-based             |
 | Grok     | `grok -p`                          | `-m` (append)            |
 | Ollama (local) | `ollama run`                  | Appended directly        |
-| Ollama (cloud) | `ollama launch claude -- --print` | `--model` (before `--`) |
+| Ollama (cloud) | `ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print` | `--model` (append) |
 
 ## Error Handling
 

commands/review.md

@@ -152,14 +152,14 @@ PROMPT_EOF
 
 **Step 4b - Run review commands in background** (run ALL in parallel with `run_in_background: true`):
 
-**Environment override for nested Claude Code**: When running inside Claude Code, `CLAUDECODE=1` prevents spawning nested sessions. Prefix with `CLAUDECODE=0` for any tool whose command starts with `claude` or uses `ollama launch claude`:
+**Environment override for nested Claude Code**: When running inside Claude Code, `CLAUDECODE=1` prevents spawning nested sessions. Prefix with `CLAUDECODE=0` for any tool whose command contains `claude`:
 
 ```bash
 # Claude tools
 cat /tmp/conclave-review-prompt.md | CLAUDECODE=0 claude --print --model opus 2>&1
 
-# Ollama cloud tools
-cat /tmp/conclave-review-prompt.md | CLAUDECODE=0 ollama launch claude --model minimax-m2.5:cloud -- --print 2>&1
+# Ollama cloud tools (runs Claude Code pointed at Ollama's API)
+cat /tmp/conclave-review-prompt.md | CLAUDECODE=0 ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print --model glm-5:cloud 2>&1
 ```
 
 For most tools (stdin-based):
@@ -181,7 +181,8 @@ For Mistral Vibe and Grok (command substitution - do not accept stdin):
 | gemini   | `-m`       | Appended to command         |
 | qwen     | `-m`       | Appended to command         |
 | mistral  | N/A        | Model set via `~/.vibe/config.toml` |
-| ollama   | varies     | See Ollama examples below   |
+| ollama (cloud) | `--model` | Appended (same as claude) |
+| ollama (local) | N/A       | Appended directly         |
 | grok     | `-m`       | Appended to command         |
 
 **Notes**:
@@ -222,10 +223,10 @@ grok -p -m grok-code-fast-1 "$(cat /tmp/conclave-review-prompt.md)"
 Original: ollama run
 With model: ollama run qwen2.5-coder:7b
 
-# Ollama cloud (--model flag before --, requires CLAUDECODE=0)
-Original: ollama launch claude -- --print
-With model: ollama launch claude --model qwen3-coder:480b-cloud -- --print
-Final: cat /tmp/prompt.md | CLAUDECODE=0 ollama launch claude --model qwen3-coder:480b-cloud -- --print 2>&1
+# Ollama cloud (--model appended, env vars point Claude at Ollama API)
+Original: ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print
+With model: ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print --model glm-5:cloud
+Final: cat /tmp/prompt.md | CLAUDECODE=0 ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print --model glm-5:cloud 2>&1
 ```
 
 Use `timeout: 300000` (5 minutes) for each command since AI tools can be slow.
@@ -653,14 +654,14 @@ Most tools receive the prompt via stdin: `cat prompt.md | {command}`
 | Mistral  | `vibe --output text -p`            | Config-based             | Uses command substitution: `vibe --output text -p "$(cat file)"` |
 | Grok     | `grok -p`                          | `-m` (append)            | Uses command substitution: `grok -p -m model "$(cat file)"` |
 | Ollama (local) | `ollama run`                  | Appended directly        | Model appended without flag: `ollama run <model>`          |
-| Ollama (cloud) | `ollama launch claude -- --print` | `--model` (before `--`) | Agentic mode with tools/web search. Requires `CLAUDECODE=0` |
+| Ollama (cloud) | `ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print` | `--model` (append) | Runs Claude Code against Ollama's API. Requires `CLAUDECODE=0` and `OLLAMA_API_KEY` |
 
 **Notes**:
 - All tools read from the same prompt file (`/tmp/conclave-review-prompt.md`) written once in Step 4a.
 - Mistral Vibe does not accept stdin; prompt must be passed via `-p` flag using command substitution.
 - Mistral model selection is done via `~/.vibe/config.toml` (`active_model` setting), not CLI flags.
 - Grok CLI does not accept stdin; prompt must be passed via `-p` flag using command substitution (like Mistral).
-- Ollama cloud models (`:cloud` suffix) use `ollama launch claude` which runs a full agentic session. The model gets access to file read, grep, bash, web search, and subagents.
+- Ollama cloud models (`:cloud` suffix) run Claude Code pointed at Ollama's Anthropic-compatible API via env vars. The model gets access to file read, grep, bash, and other Claude Code tools.
 - **Limitation**: Mistral and Grok's command-line argument passing has a ~200KB limit (ARG_MAX). Very large diffs may fail.
 
 ---

config/tools.example.json

@@ -54,58 +54,58 @@
 		"ollama-minimax": {
 			"enabled": false,
 			"scope": ["review", "consult"],
-			"command": "ollama launch claude -- --print",
+			"command": "ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print",
 			"model": "minimax-m2.5:cloud",
-			"description": "Ollama (MiniMax M2.5, agentic)"
+			"description": "Ollama (MiniMax M2.5, cloud)"
 		},
 		"ollama-glm": {
 			"enabled": false,
 			"scope": ["review", "consult"],
-			"command": "ollama launch claude -- --print",
+			"command": "ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print",
 			"model": "glm-5:cloud",
-			"description": "Ollama (GLM-5, agentic)"
+			"description": "Ollama (GLM-5, cloud)"
 		},
 		"ollama-kimi": {
 			"enabled": false,
 			"scope": ["review", "consult"],
-			"command": "ollama launch claude -- --print",
+			"command": "ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print",
 			"model": "kimi-k2.5:cloud",
-			"description": "Ollama (Kimi K2.5, agentic)"
+			"description": "Ollama (Kimi K2.5, cloud)"
 		},
 		"ollama-qwen": {
 			"enabled": false,
 			"scope": ["review", "consult"],
-			"command": "ollama launch claude -- --print",
+			"command": "ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print",
 			"model": "qwen3-coder:480b-cloud",
-			"description": "Ollama (Qwen3 Coder 480B, agentic)"
+			"description": "Ollama (Qwen3 Coder 480B, cloud)"
 		},
 		"ollama-devstral": {
 			"enabled": false,
 			"scope": ["review", "consult"],
-			"command": "ollama launch claude -- --print",
+			"command": "ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print",
 			"model": "devstral-2:123b-cloud",
-			"description": "Ollama (Devstral 2 123B, agentic)"
+			"description": "Ollama (Devstral 2 123B, cloud)"
 		},
 		"ollama-deepseek": {
 			"enabled": false,
 			"scope": ["review", "consult"],
-			"command": "ollama launch claude -- --print",
+			"command": "ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print",
 			"model": "deepseek-v3.2:cloud",
-			"description": "Ollama (DeepSeek V3.2, agentic)"
+			"description": "Ollama (DeepSeek V3.2, cloud)"
 		},
 		"ollama-rnj": {
 			"enabled": false,
 			"scope": ["review", "consult"],
-			"command": "ollama launch claude -- --print",
+			"command": "ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print",
 			"model": "rnj-1:8b-cloud",
-			"description": "Ollama (RNJ-1 8B, agentic)"
+			"description": "Ollama (RNJ-1 8B, cloud)"
 		},
 		"ollama-devstral-small": {
 			"enabled": false,
 			"scope": ["review", "consult"],
-			"command": "ollama launch claude -- --print",
+			"command": "ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com claude --print",
 			"model": "devstral-small-2:24b-cloud",
-			"description": "Ollama (Devstral Small 2 24B, agentic)"
+			"description": "Ollama (Devstral Small 2 24B, cloud)"
 		},
 		"ollama-local": {
 			"enabled": false,

tests/cli-live.sh

@@ -5,7 +5,7 @@
 # WARNING: This runs real API calls and incurs costs.
 # Only run when you need to verify tools are working end-to-end.
 #
-# NOTE: Ollama cloud models require OLLAMA_API_KEY and use `ollama launch claude` (agentic mode).
+# NOTE: Ollama cloud models require OLLAMA_API_KEY and run Claude Code via Ollama's API.
 # NOTE: Grok requires GROK_API_KEY environment variable.
 
 set -eo pipefail
@@ -100,14 +100,14 @@ else
     echo "  ○ grok skipped (GROK_API_KEY not set)"
     ((skipped++))
 fi
-# Ollama: cloud models use `ollama launch claude` (agentic), local uses `ollama run`
+# Ollama: cloud models run Claude Code via Ollama's API, local uses `ollama run`
 if [[ "$MODEL_OLLAMA" == *":cloud"* ]]; then
     if [[ -z "$OLLAMA_API_KEY" ]]; then
         echo "Testing ollama..."
         echo "  ○ ollama skipped (cloud model requires OLLAMA_API_KEY)"
         ((skipped++))
     else
-        test_tool "ollama" "ollama launch claude --model $MODEL_OLLAMA -- --print" true "CLAUDECODE=0"
+        test_tool "ollama" "claude --print --model $MODEL_OLLAMA" true "CLAUDECODE=0 ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com"
     fi
 else
     test_tool "ollama" "ollama run $MODEL_OLLAMA" true

tests/cli-models.sh

@@ -6,7 +6,7 @@
 # Only run when validating README model documentation is accurate.
 #
 # NOTE: Ollama cloud models (`:cloud` suffix) require OLLAMA_API_KEY environment variable
-# and use `ollama launch claude` (agentic mode with tools/web search).
+# and run Claude Code pointed at Ollama's Anthropic-compatible API.
 # Get your API key at https://ollama.com
 #
 # NOTE: Grok models require GROK_API_KEY environment variable.
@@ -141,13 +141,14 @@ fi
 echo ""
 
 # Ollama models
-# Cloud models use `ollama launch claude` (agentic), local models use `ollama run`
+# Cloud models run Claude Code pointed at Ollama's API, local models use `ollama run`
 echo "--- Ollama ---"
 if [[ -n "$OLLAMA_API_KEY" ]]; then
-    # Recommended cloud models (agentic via ollama launch claude)
-    test_model "ollama" "minimax-m2.5:cloud" "ollama launch claude --model minimax-m2.5:cloud -- --print" true "CLAUDECODE=0"
-    test_model "ollama" "glm-5:cloud" "ollama launch claude --model glm-5:cloud -- --print" true "CLAUDECODE=0"
-    test_model "ollama" "kimi-k2.5:cloud" "ollama launch claude --model kimi-k2.5:cloud -- --print" true "CLAUDECODE=0"
+    OLLAMA_ENV="CLAUDECODE=0 ANTHROPIC_AUTH_TOKEN=$OLLAMA_API_KEY ANTHROPIC_API_KEY= ANTHROPIC_BASE_URL=https://ollama.com"
+    # Recommended cloud models (via Ollama's Anthropic-compatible API)
+    test_model "ollama" "minimax-m2.5:cloud" "claude --print --model minimax-m2.5:cloud" true "$OLLAMA_ENV"
+    test_model "ollama" "glm-5:cloud" "claude --print --model glm-5:cloud" true "$OLLAMA_ENV"
+    test_model "ollama" "kimi-k2.5:cloud" "claude --print --model kimi-k2.5:cloud" true "$OLLAMA_ENV"
 else
     echo "  ○ cloud models skipped (OLLAMA_API_KEY not set)"
     ((skipped+=3))