ai-consultants 2.20.0 → 2.21.0

package/CLAUDE.md

@@ -6,7 +6,7 @@ AI Consultants is a multi-model AI deliberation system that queries up to 15 AI
 
 **Self-Exclusion**: The invoking agent is automatically excluded from the panel to prevent self-consultation. Claude Code won't query Claude, Codex CLI won't query Codex, etc.
 
-**Version**: 2.20.0
+**Version**: 2.21.0
 
 ## Distribution
 
@@ -38,6 +38,7 @@ ai-consultants/
 │   ├── consult_all.sh          # Main orchestrator - entry point
 │   ├── config.sh               # Centralized configuration
 │   ├── doctor.sh               # Diagnostic and auto-fix tool (v2.2)
+│   ├── update_clis.sh          # Check/update installed consultant CLIs (v2.21)
 │   ├── peer_review.sh          # Anonymous peer review (v2.2)
 │   ├── roster_audit.sh         # Roster uncorrelated-value audit (v2.20)
 │   ├── roster_calibrate.sh     # Measured capability calibration, Tier A (v2.20)
@@ -59,6 +60,7 @@ ai-consultants/
 │       ├── voting.sh           # Voting/consensus + confidence intervals
 │       ├── routing.sh          # Smart routing + cost-aware routing
 │       ├── orchestration.sh    # Dynamic orchestration planner + shapes (v2.16)
+│       ├── stance.sh           # Stance-based semantic consensus (v2.21, opt-in)
 │       ├── session.sh          # Session management
 │       ├── costs.sh            # Cost tracking + response limits
 │       ├── cache.sh            # Semantic caching (v2.3)
@@ -285,7 +287,7 @@ curl -fsSL https://ampcode.com/install.sh | bash
 ```
 
 **Environment Variables:**
-- `ENABLE_AMP` - Enable Amp consultant (default: false)
+- `ENABLE_AMP` - Enable Amp consultant (default: true)
 - `AMP_CMD` - CLI command (default: amp)
 - `AMP_TIMEOUT` - Timeout in seconds (default: 180)
 - `AMP_API_KEY` - API key for authentication
@@ -326,7 +328,7 @@ Qwen3 now supports CLI/API mode switching using the qwen-code CLI.
 export QWEN3_USE_API=false
 ./scripts/consult_all.sh "question"
 
-# API mode (default, preserves v2.6 behavior)
+# API mode (opt-in)
 export QWEN3_USE_API=true
 export QWEN3_API_KEY="your-dashscope-key"
 ./scripts/consult_all.sh "question"
@@ -342,7 +344,7 @@ npm install -g @qwen-code/qwen-code@latest
 ## v2.6 Features
 
 ### CLI/API Mode Switching
-Five consultants can now switch between CLI and API mode: **Gemini, Codex, Claude, Mistral, Qwen3**.
+Six consultants can now switch between CLI and API mode: **Gemini, Codex, Claude, Mistral, Qwen3, MiniMax**.
 
 When API mode is enabled for an agent, CLI mode is disabled (mutual exclusivity).
 
@@ -379,6 +381,7 @@ New environment variables in `config.sh`:
 | `CLAUDE_USE_API` | false | Use Anthropic API instead of claude CLI |
 | `MISTRAL_USE_API` | false | Use Mistral API instead of vibe CLI |
 | `QWEN3_USE_API` | false | Use DashScope API instead of qwen CLI (v2.7) |
+| `MINIMAX_USE_API` | false | Use MiniMax API instead of the mmx CLI (v2.21) |
 | `GEMINI_API_URL` | https://generativelanguage.googleapis.com/v1beta/models | Google AI endpoint |
 | `CODEX_API_URL` | https://api.openai.com/v1/chat/completions | OpenAI endpoint |
 | `CLAUDE_API_URL` | https://api.anthropic.com/v1/messages | Anthropic endpoint |
@@ -649,7 +652,7 @@ for f in scripts/*.sh scripts/lib/*.sh; do bash -n "$f" && echo "OK: $f"; done
 | Variable | Default | Description |
 |----------|---------|-------------|
 | `INVOKING_AGENT` | unknown | Agent invoking the skill (for self-exclusion) |
-| `ENABLE_CLAUDE` | false | Enable Claude consultant |
+| `ENABLE_CLAUDE` | true | Enable Claude consultant (auto-excluded under Claude Code) |
 | `ENABLE_DEBATE` | false | Enable Multi-Agent Debate |
 | `DEBATE_ROUNDS` | 1 | Number of debate rounds |
 | `ENABLE_SYNTHESIS` | true | Auto-synthesis of responses |
@@ -676,15 +679,19 @@ for f in scripts/*.sh scripts/lib/*.sh; do bash -n "$f" && echo "OK: $f"; done
 | `ENABLE_CAPABILITY_ROUTING` | false | Capability-aware panel composition (v2.20, opt-in) |
 | `CAPABILITY_WEIGHT_STRENGTH` | 10 | Vote-weight modulation: conf*(S+cap)/S (v2.20) |
 | `CAPABILITY_DEFAULT` | 5 | Fallback capability for a missing consultant/axis (v2.20) |
+| `ENABLE_STANCE_CONSENSUS` | false | Enumerated-stance exact-match consensus (v2.21, opt-in; +1 LLM call/run) |
+| `STANCE_MAX_OPTIONS` | 5 | Max stance options generated per question (v2.21) |
 | `ENABLE_BUDGET_LIMIT` | false | Budget enforcement (v2.4, opt-in) |
 | `BUDGET_ACTION` | warn | Action on budget exceeded: warn/stop (v2.4) |
 | `QWEN3_USE_API` | false | Use DashScope API instead of qwen CLI (v2.7) |
 | `QWEN3_CMD` | qwen | Qwen CLI command (v2.7) |
-| `ENABLE_KIMI` | false | Enable Kimi consultant (v2.9) |
+| `ENABLE_KIMI` | true | Enable Kimi consultant (v2.9) |
 | `KIMI_CMD` | kimi | Kimi CLI command (v2.9) |
 | `KIMI_MODEL` | kimi-code/kimi-for-coding | Kimi model (v2.9) |
-| `ENABLE_MINIMAX` | false | Enable MiniMax consultant (v2.10) |
-| `MINIMAX_API_KEY` | - | MiniMax API key (v2.10) |
+| `ENABLE_MINIMAX` | true | Enable MiniMax consultant (v2.10; CLI via mmx v2.21) |
+| `MINIMAX_USE_API` | false | Use MiniMax API instead of the mmx CLI (v2.21) |
+| `MINIMAX_CMD` | mmx | MiniMax CLI command (v2.21) |
+| `MINIMAX_API_KEY` | - | MiniMax API key (API mode only) (v2.10) |
 | `MINIMAX_MODEL` | MiniMax-M2.7 | MiniMax model (v2.10) |
 
 ## External Dependencies
@@ -699,6 +706,7 @@ for f in scripts/*.sh scripts/lib/*.sh; do bash -n "$f" && echo "OK: $f"; done
 - `kimi` CLI - Kimi Code (v2.9)
 - `claude` CLI - Claude (v2.2, also used for synthesis)
 - `qwen` CLI - Qwen via qwen-code (v2.7, optional)
+- `mmx` CLI - MiniMax via mmx-cli (v2.21, optional; `npm i -g mmx-cli`, auth `mmx auth login`)
 - `ollama` CLI - Local models (v2.2)
 - `jq` - JSON parsing
 
@@ -834,6 +842,15 @@ curl -fsSL https://raw.githubusercontent.com/matteoscurati/ai-consultants/main/s
 
 ## Changelog
 
+### v2.21.0
+- **CLI-first transport (principle).** Every `*_USE_API` defaults `false`; a consultant with a CLI always uses it, API is opt-in (CLI-less model or explicit choice). `ENABLE_AMP`/`ENABLE_CLAUDE` default true.
+- **MiniMax via `mmx` CLI** (was API-only). `query_minimax.sh` runs `mmx text chat --non-interactive …` in CLI mode; `config.sh` adds `MINIMAX_CMD` + auto-resolves `MINIMAX_USE_API` (API iff `MINIMAX_API_KEY` set — back-compat), drops MiniMax from `API_CONSULTANTS`; `doctor.sh` treats it as switchable.
+- **`scripts/update_clis.sh`** (+ `ai-consultants update-clis`): per CLI-backed consultant, `detect_method` (brew formula/cask, npm, uv, pipx, pip, self-update, curl installer) → update. `--dry-run`, `--only`.
+- **Stance consensus** (opt-in, `ENABLE_STANCE_CONSENSUS`): new `lib/stance.sh` (`generate_stance_options` — one synthesizer call wrapped in `run_with_timeout`; `_stance_clean`; `build_stance_prompt`). `consult_all.sh` generates the shared set → `stance_options.json` + exports `STANCE_OPTIONS_PROMPT`; injected via `personas.sh` (CLI) + `api_query.sh` (API). `voting.sh::calculate_consensus_score` scores the plurality stance over the PANEL size (not the count of emitted stances). `STANCE_MAX_OPTIONS`, `STANCE_TIMEOUT`.
+- **Smoke-test reliability**: `query_gemini.sh` (`agy -p "$QUERY"`), `query_kimi.sh` (`stream-json` + `_kimi_extract_content`), `query_kilo.sh` (de-fence). Shared `_is_consultant_response_file` metadata filter in `common.sh`, applied across `voting.sh`/`synthesize.sh`/`peer_review.sh`/`reflection.sh`. `orchestration.sh` stalled→stable relabel (set -e-safe) + debate-round promote validation.
+- **Persona fix**: `_AGENT_DEFAULT_PERSONAS` mapped `GLM|17`/`DEEPSEEK|7` (each got the other's persona) → `GLM|7`/`DEEPSEEK|17`.
+- **Dev/reliability**: reliability tracking + Amp/Claude default-on; offline e2e test; `scripts/release.sh` version-bump automation; calibration fixes (diagnosable peer-review failures + cost-only path); `test_user_config.sh` XDG test made hermetic (unset the exported `_AI_CONSULTANTS_XDG_*` intermediates). `npm test` = 18 suites. Hardened by a `/code-review max` pass.
+
 ### v2.20.0
 - **Capability axes (borrowed from the delegation-kit cost/intelligence/taste table).** `references/affinity.json` → v1.1: new `capabilities` (per-consultant {intelligence, taste, cost}, 1-10), `category_axis` (category → the quality axis it stresses: taste for API_DESIGN/ARCHITECTURE/CODE_REVIEW/GENERAL, intelligence otherwise), `capability_default`. New `lib/routing.sh::get_capability` / `get_category_axis` (cached like `get_affinity`).
 - **Capability-weighted voting** (`ENABLE_CAPABILITY_WEIGHTING`, opt-in): `lib/voting.sh::_effective_vote_weight` modulates a vote by the consultant's capability on the run's axis — `confidence × (S + cap) / S` (`S = CAPABILITY_WEIGHT_STRENGTH`, default 10). Applied in `calculate_weighted_recommendation` and `calculate_final_score` (normalizer uses the same weight → 1-10 scale preserved). Axis from `QUESTION_CATEGORY`. `cost` never weights a vote (tie-break intelligence > taste > cost).

package/README.md

@@ -1,8 +1,8 @@
-# AI Consultants v2.20.0
+# AI Consultants v2.21.0
 
 > **A harness for every question.** A panel of up to 15 frontier models that writes its own playbook per question — fan out, debate to convergence, cross-examine under adversarial review, or run a tournament — and checks its work before it reaches you.
 
-[![Version](https://img.shields.io/badge/version-2.20.0-blue.svg)](https://github.com/matteoscurati/ai-consultants)
+[![Version](https://img.shields.io/badge/version-2.21.0-blue.svg)](https://github.com/matteoscurati/ai-consultants)
 [![License](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)
 [![Claude Code Skill](https://img.shields.io/badge/Claude%20Code-Skill-orange.svg)](https://docs.anthropic.com/en/docs/claude-code/skills)
 [![GitHub stars](https://img.shields.io/github/stars/matteoscurati/ai-consultants?style=social)](https://github.com/matteoscurati/ai-consultants)
@@ -580,6 +580,8 @@ ai-consultants doctor --json                                   # JSON for automa
 ai-consultants doctor --live                                   # Real ping per consultant — catches installed-but-unauthenticated CLIs
 ai-consultants doctor --suggest-config                         # Print recommended ENABLE_* based on detected CLIs
 ai-consultants doctor --suggest-preset --question "..."        # Recommend preset + strategy for a question
+ai-consultants update-clis                                     # Check & update every installed consultant CLI
+ai-consultants update-clis --dry-run                           # Preview: each CLI's install method + update command
 ```
 
 > When a consultant fails mid-consultation, the run surfaces the captured reason (e.g. `CLI not found`, `401 Unauthorized`) instead of a bare "Failed", so you can tell *not installed* from *not authenticated* from *transient*.
@@ -609,6 +611,11 @@ The **shape** is chosen per question (or pinned via `ORCHESTRATION_MODE`):
 
 Set `ORCHESTRATION_MODE=fixed` for the classic fixed-round pipeline.
 
+Consensus is measured lexically by default; set `ENABLE_STANCE_CONSENSUS=true`
+(opt-in, v2.21) to have the panel pick from a shared set of enumerated stance
+options and score agreement by exact match instead — see
+[references/configuration.md](references/configuration.md#semantic-consensus-v221-opt-in).
+
 With peer review:
 ```
 Responses -> Anonymize -> Peer Ranking -> De-anonymize -> Peer Scores

package/SKILL.md

@@ -5,10 +5,10 @@ license: MIT
 compatibility: Requires bash, jq, and at least 2 AI CLI tools (agy, codex, vibe, etc.). macOS and Linux.
 metadata:
   author: matteoscurati
-  version: 2.20.0
+  version: 2.21.0
 ---
 
-# AI Consultants v2.20.0 - AI Expert Panel
+# AI Consultants v2.21.0 - AI Expert Panel
 
 **A harness for every question.** Convene a panel of AI "consultants" for coding questions — and let the engine pick *how* they deliberate: a quick read, a convergence loop, an adversarial refutation gate, a tournament, or an exhaustive sweep. Each consultant has a **configurable persona** that shapes its analysis.
 
@@ -42,11 +42,12 @@ Configuration (presets, strategies, features, personas, API keys) can be managed
 | **Kimi** | `kimi` | The Eastern Sage | Holistic, balanced perspectives |
 | **Claude** | `claude` | The Synthesizer | Big picture, synthesis |
 | **Qwen** | `qwen` | The Analyst | Data-driven, metrics |
+| **MiniMax** | `mmx` | The Pragmatic Optimizer | Performance, efficiency, pragmatism |
 | **Ollama** | `ollama` | The Local Expert | Privacy-first, zero cost |
 
-**API-only consultants**: GLM (The Methodologist), Grok (The Provocateur), DeepSeek (The Code Specialist), MiniMax (The Pragmatic Optimizer)
+**API-only consultants**: GLM (The Methodologist), Grok (The Provocateur), DeepSeek (The Code Specialist)
 
-**CLI/API Mode**: Gemini, Codex, Claude, Mistral, and Qwen can switch between CLI and API mode via `*_USE_API` environment variables. Gemini auto-selects API mode when `GEMINI_API_KEY` is set (no `agy` install needed) and the CLI otherwise.
+**CLI/API Mode**: Gemini, Codex, Claude, Mistral, Qwen, and MiniMax can switch between CLI and API mode via `*_USE_API` environment variables. Gemini auto-selects API mode when `GEMINI_API_KEY` is set (no `agy` install needed) and the CLI otherwise.
 
 **Self-Exclusion**: The invoking agent is automatically excluded from the panel. When invoked from Claude Code, Claude is excluded; when invoked from Codex CLI, Codex is excluded, etc.
 
@@ -96,6 +97,7 @@ Query is classified, then sent to consultants in parallel. Responses are scored,
 | **Smart Routing** | Auto-select best consultants per question type | `ENABLE_SMART_ROUTING` |
 | **Cost Tracking** | Track API usage costs | `ENABLE_COST_TRACKING` |
 | **Panic Mode** | Auto-add rigor when uncertainty detected | `ENABLE_PANIC_MODE` |
+| **Stance Consensus** | Panel picks from enumerated stances; agreement by exact match (opt-in, v2.21) | `ENABLE_STANCE_CONSENSUS` |
 
 ## Configuration
 
@@ -136,6 +138,10 @@ Diagnose and fix issues:
 ./scripts/doctor.sh --live       # Real ping per consultant — catches installed-but-unauthenticated CLIs
 ```
 
+Keep the consultant CLIs current with `ai-consultants update-clis` (`--dry-run` to
+preview, `--only <cli>` to target one) — it detects how each CLI was installed
+(brew/npm/uv/pipx/curl/self-update) and updates it.
+
 When a consultant fails during a consultation, the run prints the captured reason (e.g. "CLI not found", "401 Unauthorized"). The run is also **graded** MET/DEGRADED/FAILED by how many responded (`QUORUM_MIN`), with a "Diagnosed Failures" section in the report. Set `ENABLE_HEALTH_GATE=true` to ping consultants and drop the dead ones *before* the run.
 
 ## Interpreting Results

package/bin/ai-consultants

@@ -55,6 +55,7 @@ Usage:
   ai-consultants --preset <name> "question"   Use a configuration preset
   ai-consultants --strategy <name> "question" Use a synthesis strategy
   ai-consultants doctor [--fix] [--json]      Run diagnostics
+  ai-consultants update-clis [--dry-run]      Check & update installed consultant CLIs
   ai-consultants init [--force]               Scaffold ~/.config/ai-consultants/{config.sh,.env}
   ai-consultants install                      Install slash commands to ~/.claude/commands/
   ai-consultants version                      Show version
@@ -78,6 +79,10 @@ EOF
     shift
     exec "$SCRIPTS_DIR/doctor.sh" "$@"
     ;;
+  update-clis|update)
+    shift
+    exec "$SCRIPTS_DIR/update_clis.sh" "$@"
+    ;;
   init)
     shift
     FORCE=false

package/docs/COST_RATES.md

@@ -1,4 +1,4 @@
-# Cost Rates - AI Consultants v2.20.0
+# Cost Rates - AI Consultants v2.21.0
 
 This page documents the per-token rates used by the cost tracking system.
 

package/docs/SETUP.md

@@ -358,15 +358,19 @@ curl -L code.kimi.com/install.sh | bash
 
 Verify: `kimi --version`
 
-### MiniMax (API-only) - v2.10
+### MiniMax (CLI via mmx) - v2.21
 
-MiniMax requires an API key — no CLI installation needed.
+MiniMax runs via the `mmx` CLI by default (CLI-first, OAuth). Install and sign in:
 
 ```bash
-export MINIMAX_API_KEY="your-key"
-export ENABLE_MINIMAX=true
+npm install -g mmx-cli
+mmx auth login          # OAuth (browser); or: mmx auth login --api-key <key>
 ```
 
+API mode is opt-in: set `MINIMAX_USE_API=true` with `MINIMAX_API_KEY`.
+
+Verify: `mmx --version`
+
 ---
 
 ## Ollama (Local Models) - v2.2
@@ -521,7 +525,7 @@ INVOKING_AGENT=codex ./scripts/consult_all.sh "Question"
 
 ## CLI/API Mode Switching (v2.6+)
 
-Five consultants can switch between CLI mode (using local CLI tools) and API mode (direct API calls): **Gemini, Codex, Claude, Mistral, and Qwen3**.
+Six consultants can switch between CLI mode (using local CLI tools) and API mode (direct API calls): **Gemini, Codex, Claude, Mistral, Qwen3, and MiniMax**.
 
 ### Why Use API Mode?
 

package/docs/cost_rates.json

@@ -1,5 +1,5 @@
 {
-  "version": "2.20.0",
+  "version": "2.21.0",
   "updated": "2026-06-24",
   "models": {
     "_comment_units": "All rates are USD per 1,000 tokens (per-1K). lib/costs.sh computes cost as (token_count / 1000) * rate, so every value here MUST be a per-1K rate = (provider $/1M price) / 1000. Do NOT paste $/MTok figures directly.",

package/docs/releases/v2.21.0.md

@@ -0,0 +1,65 @@
+# Release v2.21.0
+
+**Date:** 2026-07-14
+**Type:** Minor — CLI-first transport, MiniMax on the mmx CLI, `update-clis`, and opt-in stance consensus
+**Previous:** v2.20.0
+
+## Highlights
+
+- **CLI-first is now the rule.** Any consultant that ships a CLI uses it by default; API mode is opt-in — only for CLI-less models or an explicit `*_USE_API=true`. Amp and Claude are on by default.
+- **MiniMax runs on the `mmx` CLI** (it was API-only). Existing API-only setups keep working: `MINIMAX_USE_API` auto-resolves to API mode only when `MINIMAX_API_KEY` is set.
+- **`ai-consultants update-clis`** keeps every consultant CLI current, detecting how each was installed.
+- **Stance-based semantic consensus** (opt-in): the panel picks from a shared set of enumerated stance options, so agreement is scored by exact match instead of lexical similarity.
+- A batch of smoke-test-surfaced reliability fixes across the Gemini/Kimi/Kilo adapters and the voting/synthesis pipeline, plus a runtime persona fix. Hardened by a multi-agent `/code-review max` pass.
+
+## What's New
+
+### CLI-first transport
+
+Every `*_USE_API` flag now defaults to `false`. A consultant with a CLI always uses it; API mode is reserved for models without a CLI or an explicit opt-in. `ENABLE_AMP` and `ENABLE_CLAUDE` default on.
+
+### MiniMax via the `mmx` CLI
+
+`query_minimax.sh` runs `mmx text chat --non-interactive --quiet --no-color --output text …` by default. `MINIMAX_CMD` (default `mmx`) sets the binary. To preserve pre-2.21 behavior, `MINIMAX_USE_API` auto-resolves:
+
+- `MINIMAX_API_KEY` set → API mode (the only transport before 2.21 — nothing breaks on upgrade).
+- otherwise → the `mmx` CLI.
+
+An explicit `MINIMAX_USE_API=true|false` always wins. `doctor` validates the CLI in CLI mode and the key in API mode.
+
+### `update-clis`
+
+```bash
+ai-consultants update-clis            # update every installed consultant CLI
+ai-consultants update-clis --dry-run  # show each CLI's install method + the exact update command
+ai-consultants update-clis --only mmx # limit to one CLI (by consultant name or binary)
+```
+
+For each CLI-backed consultant it detects the install method — brew formula/cask, npm, uv, pipx, pip, a self-updating binary, or a curl installer — and updates it via the matching command, best-effort.
+
+### Stance-based semantic consensus (opt-in)
+
+| Flag (default) | Effect |
+|---|---|
+| `ENABLE_STANCE_CONSENSUS` (false) | One LLM call enumerates 2–`STANCE_MAX_OPTIONS` mutually-exclusive stance options for the question; each consultant picks one verbatim; consensus = the plurality stance's share of the **panel**. |
+| `STANCE_MAX_OPTIONS` (5) | Cap on the generated option set. |
+| `STANCE_TIMEOUT` (60) | Seconds for the stance-generation call (guards a blocking/unauthenticated synthesizer at startup). |
+
+The lexical cluster (Jaccard over the free-text `approach`) can't tell that "Commit the lockfile" and "Always keep package-lock in git" agree; the stance signal makes agreement exact-matchable. Every path degrades to the lexical cluster if fewer than two stances are emitted or generation fails, so it is always safe to enable. The generated options are written to `stance_options.json` in the output dir.
+
+### Reliability fixes
+
+- **CLI adapters** (from a full-panel smoke test): Gemini (`agy -p "$QUERY"`, not the broken `-p -`), Kimi (`--output-format stream-json` + a hardened extractor), and Kilo (strip the ```json fence) return structured responses instead of the unstructured fallback.
+- **Metadata pollution**: a shared `_is_consultant_response_file` filter keeps pipeline files (`voting.json`, `orchestration.json`, `stance_options.json`, …) out of every voting/synthesis/peer-review glob, so they can't become phantom votes, reviewees, or synthesis inputs. Consensus is the largest agreeing cluster (single-linkage over Jaccard).
+- **Orchestration**: a lexical "stalled" is relabeled "stable" when the panel's own per-round signal says positions stopped moving (now `set -e`-safe against a malformed round summary); a debate round's refined stance is validated before it is grafted.
+- **Personas**: GLM and DeepSeek were transposed at runtime — each consultant ran under the other's persona name and prompt. Fixed.
+
+### Under the hood
+
+Reliability tracking with Amp/Claude default-on; an offline end-to-end integration test; `scripts/release.sh` version-bump automation; calibration fixes (diagnosable peer-review failures + a cost-only path). `npm test` now runs 18 suites.
+
+## Upgrade notes
+
+- **Fully back-compat.** With the new flags off, behavior matches v2.20.0.
+- **MiniMax users**: if you ran MiniMax via `MINIMAX_API_KEY`, nothing changes — it stays on the API. To switch to the CLI, `npm install -g mmx-cli` and `mmx auth login`, then unset the key (or set `MINIMAX_USE_API=false`).
+- **New opt-in flags**: `ENABLE_STANCE_CONSENSUS`, `STANCE_MAX_OPTIONS`, `STANCE_TIMEOUT`.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ai-consultants",
-  "version": "2.20.0",
+  "version": "2.21.0",
   "description": "A harness for every question - a panel of up to 15 AI models that picks its own deliberation strategy per question (convergence, adversarial verification, tournament) and checks its work before answering",
   "bin": {
     "ai-consultants": "bin/ai-consultants"

package/references/configuration.md

@@ -141,9 +141,32 @@ agrees on the first pass.
 The convergence trajectory and stop reason are recorded in `orchestration.json` /
 `optimization_metrics.json`. Every round still respects `MAX_SESSION_COST` / budget limits.
 
+## Semantic Consensus (v2.21+, opt-in)
+
+By default the consensus score (which drives `CONVERGENCE_TARGET_CONSENSUS`) is a
+**lexical cluster**: the largest group of consultants whose free-text `approach`
+fields are similar (single-linkage over Jaccard). That can't tell that two
+differently-phrased answers actually AGREE ("Commit the lockfile" vs "Always keep
+package-lock in git").
+
+`ENABLE_STANCE_CONSENSUS=true` adds an exact-matchable signal: one extra LLM call
+enumerates a small set of mutually-exclusive **stance options** for the question,
+each consultant is asked to pick exactly one verbatim, and consensus becomes the
+plurality stance's share of the **panel** (consultants that answered but emitted no
+stance count against agreement). It degrades to the lexical cluster whenever fewer
+than two stances are emitted or generation fails, so it is always safe to enable.
+
+```bash
+ENABLE_STANCE_CONSENSUS=false   # opt-in; adds ~1 LLM call per run
+STANCE_MAX_OPTIONS=5            # max stance options generated per question
+STANCE_TIMEOUT=60              # seconds for the stance-generation call (guards a hang)
+```
+
+The generated options are written to `stance_options.json` in the output dir.
+
 ## CLI/API Mode Switching (v2.6+)
 
-Five consultants support switching between CLI and API mode. When API mode is enabled, the CLI is not used.
+Six consultants support switching between CLI and API mode. **The default is the CLI** for every consultant that has one — API mode is opt-in (for CLI-less models or an explicit choice). When API mode is enabled, the CLI is not used.
 
 **Gemini auto-resolution (v2.15.1):** leave `GEMINI_USE_API` unset and the mode is chosen for you — API mode when `GEMINI_API_KEY` is present (no `agy` install or OAuth needed; ideal for `npx`), CLI mode (`agy`) otherwise. Set `GEMINI_USE_API` explicitly only to force a mode (an explicit value disables auto-detection).
 
@@ -154,6 +177,7 @@ CODEX_USE_API=false          # Use OpenAI API instead of codex CLI
 CLAUDE_USE_API=false         # Use Anthropic API instead of claude CLI
 MISTRAL_USE_API=false        # Use Mistral API instead of vibe CLI
 QWEN3_USE_API=false          # Use qwen CLI (default) or DashScope API
+MINIMAX_USE_API=false        # Use mmx CLI (default) or MiniMax API
 ```
 
 ### API Keys
@@ -165,6 +189,7 @@ QWEN3_USE_API=false          # Use qwen CLI (default) or DashScope API
 | Claude | `ANTHROPIC_API_KEY` | Anthropic API key |
 | Mistral | `MISTRAL_API_KEY` | Mistral API key |
 | Qwen3 | `QWEN3_API_KEY` | DashScope API key |
+| MiniMax | `MINIMAX_API_KEY` | MiniMax API key (API mode only; the mmx CLI uses OAuth) |
 
 ## Consultant Toggles and Models
 
@@ -175,17 +200,17 @@ ENABLE_CODEX=true
 ENABLE_MISTRAL=true
 ENABLE_KILO=true
 ENABLE_CURSOR=true
-ENABLE_AIDER=true
-
-# Opt-in consultants
-ENABLE_AMP=false             # Amp CLI - The Systems Thinker
-ENABLE_KIMI=false            # Kimi CLI - The Eastern Sage
-ENABLE_QWEN3=false           # Qwen CLI/API - The Analyst
-ENABLE_CLAUDE=false          # Claude CLI - The Synthesizer
+ENABLE_AMP=true              # Amp CLI - The Systems Thinker
+ENABLE_KIMI=true             # Kimi CLI - The Eastern Sage
+ENABLE_QWEN3=true            # Qwen CLI/API - The Analyst
+ENABLE_MINIMAX=true          # MiniMax CLI/API (mmx) - The Pragmatic Optimizer
+ENABLE_CLAUDE=true           # Claude CLI - The Synthesizer (auto-excluded under Claude Code)
+
+# Off by default
+ENABLE_AIDER=false           # Aider CLI - The Pair Programmer
 ENABLE_OLLAMA=false          # Ollama - The Local Expert
-ENABLE_MINIMAX=false         # MiniMax API - The Pragmatic Optimizer
 
-# API-only consultants
+# API-only consultants (off by default - require API keys)
 ENABLE_GLM=false
 ENABLE_GROK=false
 ENABLE_DEEPSEEK=false

package/scripts/config.sh

@@ -64,9 +64,13 @@ DEFAULT_OUTPUT_DIR_BASE="${DEFAULT_OUTPUT_DIR_BASE:-${_AI_CONSULTANTS_XDG_CACHE}
 # =============================================================================
 # For agents that support both CLI and API mode, set USE_API=true to use API mode.
 # When API mode is enabled, CLI mode is automatically disabled (mutual exclusivity).
-# 5 agents support switching: Gemini, Codex, Claude, Mistral, Qwen3
+# 6 agents support switching: Gemini, Codex, Claude, Mistral, Qwen3, MiniMax
 
-# Mode switching (true = use API, false = use CLI)
+# Mode switching (true = API, false = CLI). DEFAULT IS CLI — when a consultant
+# has a CLI, the tool uses it; the CLIs are the primary transport (OAuth /
+# subscription, no API key needed). API mode is opt-in only: for CLI-less models
+# (GLM/Grok/DeepSeek are API-only anyway) or an explicit per-run user
+# choice. Set a switch to true (with its API key) to force API for that agent.
 # GEMINI_USE_API is intentionally NOT defaulted here: it is auto-resolved in the
 # Gemini configuration section below (needs GEMINI_API_KEY / GEMINI_CMD), so an
 # explicit unset must remain distinguishable from an explicit "false".
@@ -74,6 +78,10 @@ CODEX_USE_API="${CODEX_USE_API:-false}"
 CLAUDE_USE_API="${CLAUDE_USE_API:-false}"
 MISTRAL_USE_API="${MISTRAL_USE_API:-false}"
 QWEN3_USE_API="${QWEN3_USE_API:-false}"  # Default false to use qwen CLI
+# MINIMAX_USE_API is intentionally NOT defaulted here: it is auto-resolved in the
+# MiniMax configuration section below (needs MINIMAX_API_KEY), so an explicit
+# unset stays distinguishable from an explicit "false" (back-compat for pre-v2.21
+# API-only MiniMax users -- see the rationale at the resolve block).
 
 # API endpoints for CLI-switchable agents
 GEMINI_API_URL="${GEMINI_API_URL:-https://generativelanguage.googleapis.com/v1beta/models}"
@@ -215,14 +223,36 @@ DEEPSEEK_FORMAT="${DEEPSEEK_FORMAT:-openai}"
 # API key: Set DEEPSEEK_API_KEY environment variable
 
 # =============================================================================
-# MINIMAX CONFIGURATION - The Pragmatic Optimizer (API-based, v2.10)
+# MINIMAX CONFIGURATION - The Pragmatic Optimizer (CLI/API, v2.10; CLI via mmx v2.21)
 # =============================================================================
 
+MINIMAX_CMD="${MINIMAX_CMD:-mmx}"
 MINIMAX_MODEL="${MINIMAX_MODEL:-MiniMax-M2.7}"
 MINIMAX_TIMEOUT_SECONDS="${MINIMAX_TIMEOUT:-180}"
 MINIMAX_API_URL="${MINIMAX_API_URL:-https://api.minimax.io/v1/chat/completions}"
 MINIMAX_FORMAT="${MINIMAX_FORMAT:-openai}"
-# API key: Set MINIMAX_API_KEY environment variable
+# CLI mode (default) uses the mmx CLI (npm i -g mmx-cli; auth: mmx auth login).
+# API mode (MINIMAX_USE_API=true) uses MINIMAX_API_KEY against MINIMAX_API_URL.
+#
+# Auto-resolve the MiniMax transport when the user hasn't pinned MINIMAX_USE_API.
+# Unlike Gemini, the mmx CLI IS npm-installable, so CLI is the genuine default
+# (per the CLI-first principle). This block exists purely for BACK-COMPAT: before
+# v2.21 MiniMax was API-only, so a working pre-v2.21 config necessarily had
+# MINIMAX_API_KEY and no mmx. Defaulting such a user to CLI would break MiniMax on
+# upgrade (mmx not found). So when the mode is unset:
+#   - MINIMAX_API_KEY present -> API mode (preserves the pre-v2.21 API-only user;
+#                                a set key is itself the "explicit user choice")
+#   - otherwise               -> CLI mode (mmx; the new default)
+# An explicit MINIMAX_USE_API=true/false is always honored. Idempotent across
+# re-sourcing via the "${MINIMAX_USE_API+x}" set-vs-unset guard.
+if [[ -z "${MINIMAX_USE_API+x}" ]]; then
+    if [[ -n "${MINIMAX_API_KEY:-}" ]]; then
+        MINIMAX_USE_API=true
+    else
+        MINIMAX_USE_API=false
+    fi
+fi
+export MINIMAX_USE_API
 
 # =============================================================================
 # AMP CONFIGURATION - The Systems Thinker (v2.8)
@@ -285,7 +315,7 @@ ALL_CONSULTANTS=("Gemini" "Codex" "Mistral" "Kilo" "Cursor" "Aider" "Amp" "Kimi"
 CLI_CONSULTANTS=("Gemini" "Codex" "Mistral" "Kilo" "Cursor" "Aider" "Amp" "Kimi" "Claude" "Qwen3" "Ollama")
 
 # API-only consultants (use HTTP API directly, no CLI available)
-API_CONSULTANTS=("GLM" "Grok" "DeepSeek" "MiniMax")
+API_CONSULTANTS=("GLM" "Grok" "DeepSeek")
 
 # =============================================================================
 # ENABLED CONSULTANTS
@@ -298,17 +328,17 @@ ENABLE_CODEX="${ENABLE_CODEX:-true}"
 ENABLE_MISTRAL="${ENABLE_MISTRAL:-true}"
 ENABLE_KILO="${ENABLE_KILO:-true}"
 ENABLE_CURSOR="${ENABLE_CURSOR:-true}"
-ENABLE_AIDER="${ENABLE_AIDER:-false}"
-ENABLE_AMP="${ENABLE_AMP:-false}"        # Amp Code (v2.8)
+ENABLE_AMP="${ENABLE_AMP:-true}"         # Amp Code (v2.8)
 ENABLE_KIMI="${ENABLE_KIMI:-true}"       # Kimi Code (v2.9)
-ENABLE_CLAUDE="${ENABLE_CLAUDE:-false}"  # Auto-disabled when invoked by Claude Code
+ENABLE_CLAUDE="${ENABLE_CLAUDE:-true}"   # Auto-disabled when invoked by Claude Code
+ENABLE_QWEN3="${ENABLE_QWEN3:-true}"     # qwen-code CLI (v2.7); API opt-in
+ENABLE_MINIMAX="${ENABLE_MINIMAX:-true}" # mmx CLI (v2.21); API opt-in
+ENABLE_AIDER="${ENABLE_AIDER:-false}"    # Aider CLI (off by default)
 
-# API-based consultants (disabled by default - require API keys)
-ENABLE_QWEN3="${ENABLE_QWEN3:-true}"
+# API-only consultants (disabled by default - require API keys)
 ENABLE_GLM="${ENABLE_GLM:-false}"
 ENABLE_GROK="${ENABLE_GROK:-false}"
 ENABLE_DEEPSEEK="${ENABLE_DEEPSEEK:-false}"
-ENABLE_MINIMAX="${ENABLE_MINIMAX:-false}"
 
 # Local model support via Ollama (disabled by default)
 ENABLE_OLLAMA="${ENABLE_OLLAMA:-false}"
@@ -440,6 +470,13 @@ WARN_AT_COST="${WARN_AT_COST:-0.50}"
 # (persistent across reboots; this is user data, not cache).
 COST_TRACKING_FILE="${COST_TRACKING_FILE:-${_AI_CONSULTANTS_XDG_DATA}/costs.json}"
 
+# --- Reliability Tracking (foundation for a future self-tuning roster) ---
+# Enable per-consultant success/failure recording (opt-in default: ON)
+ENABLE_RELIABILITY_TRACKING="${ENABLE_RELIABILITY_TRACKING:-true}"
+
+# File for cumulative per-consultant reliability tracking (same XDG data dir as costs)
+RELIABILITY_FILE="${RELIABILITY_FILE:-${_AI_CONSULTANTS_XDG_DATA}/reliability.json}"
+
 # --- Budget Enforcement (v2.4) ---
 # Enable budget limit enforcement (opt-in, default OFF)
 ENABLE_BUDGET_LIMIT="${ENABLE_BUDGET_LIMIT:-false}"
@@ -586,6 +623,15 @@ CONVERGENCE_STALL_EPSILON="${CONVERGENCE_STALL_EPSILON:-5}"
 # and runs anonymous peer review as a refutation gate before synthesis.
 ENABLE_ADVERSARIAL_VERIFY="${ENABLE_ADVERSARIAL_VERIFY:-true}"
 
+# Stance-based semantic consensus (v2.21, opt-in). When true, one LLM call
+# enumerates 2..STANCE_MAX_OPTIONS mutually-exclusive stance options per question
+# and each consultant picks one verbatim, so consensus is an exact-match plurality
+# (robust to paraphrasing) instead of lexical clustering. Adds one serial LLM call
+# per run; degrades gracefully to the cluster consensus if generation fails.
+ENABLE_STANCE_CONSENSUS="${ENABLE_STANCE_CONSENSUS:-false}"
+STANCE_MAX_OPTIONS="${STANCE_MAX_OPTIONS:-5}"
+STANCE_TIMEOUT="${STANCE_TIMEOUT:-60}"   # seconds for the stance-generation call (guards a hang)
+
 # =============================================================================
 # QUORUM GRADING (v2.19.0)
 # =============================================================================
@@ -899,4 +945,4 @@ EOF
 # VERSION
 # =============================================================================
 
-AI_CONSULTANTS_VERSION="2.20.0"
+AI_CONSULTANTS_VERSION="2.21.0"

package/scripts/consult_all.sh

@@ -36,6 +36,7 @@ source "$SCRIPT_DIR/lib/common.sh"
 source "$SCRIPT_DIR/lib/session.sh"
 source "$SCRIPT_DIR/lib/progress.sh"
 source "$SCRIPT_DIR/lib/costs.sh"
+source "$SCRIPT_DIR/lib/reliability.sh"
 source "$SCRIPT_DIR/lib/voting.sh"
 source "$SCRIPT_DIR/lib/routing.sh"
 source "$SCRIPT_DIR/lib/cache.sh"
@@ -298,6 +299,21 @@ chmod 700 "$DEFAULT_OUTPUT_DIR_BASE"
 mkdir -p "$OUTPUT_DIR"
 chmod 700 "$OUTPUT_DIR"
 
+# --- Stance options (v2.21, opt-in): generate a shared enumerated stance set so
+# every consultant picks the same one -> exact-match consensus. Fully graceful:
+# on any failure STANCE_OPTIONS_PROMPT stays empty and consensus uses the cluster.
+if [[ "${ENABLE_STANCE_CONSENSUS:-false}" == "true" ]]; then
+    source "$SCRIPT_DIR/lib/stance.sh"
+    STANCE_OPTIONS=$(generate_stance_options "$QUERY" 2>/dev/null || echo "[]")
+    echo "$STANCE_OPTIONS" > "$OUTPUT_DIR/stance_options.json" 2>/dev/null || true
+    if STANCE_OPTIONS_PROMPT=$(build_stance_prompt "$STANCE_OPTIONS" 2>/dev/null); then
+        export STANCE_OPTIONS_PROMPT
+        log_info "Stance consensus: $(echo "$STANCE_OPTIONS" | jq -r 'length' 2>/dev/null || echo '?') options generated"
+    else
+        log_warn "Stance consensus: no usable options — falling back to lexical cluster consensus"
+    fi
+fi
+
 log_info "Output: $OUTPUT_DIR"
 log_info "Question: ${QUERY:0:80}$([ ${#QUERY} -gt 80 ] && echo '...')"
 if [[ ${#FILES[@]} -gt 0 ]]; then
@@ -557,9 +573,15 @@ for i in "${!PIDS[@]}"; do
             log_success "  $name: OK (cached)"
             RESULTS+=("$name:OK")
             SUCCESS_COUNT=$((SUCCESS_COUNT + 1))
+            if [[ "${ENABLE_RELIABILITY_TRACKING:-true}" == "true" ]]; then
+                record_consultant_outcome "$name" success 2>/dev/null || true
+            fi
         else
             log_warn "  $name: Cache miss"
             RESULTS+=("$name:EMPTY")
+            if [[ "${ENABLE_RELIABILITY_TRACKING:-true}" == "true" ]]; then
+                record_consultant_outcome "$name" fail 2>/dev/null || true
+            fi
         fi
         continue
     fi
@@ -570,6 +592,9 @@ for i in "${!PIDS[@]}"; do
             RESULTS+=("$name:OK")
             SUCCESS_COUNT=$((SUCCESS_COUNT + 1))
             update_progress "$name" 100 "success"
+            if [[ "${ENABLE_RELIABILITY_TRACKING:-true}" == "true" ]]; then
+                record_consultant_outcome "$name" success 2>/dev/null || true
+            fi
 
             # Store in cache (v2.3)
             if is_cache_enabled; then
@@ -581,12 +606,18 @@ for i in "${!PIDS[@]}"; do
             _surface_consultant_error "$name" "$output_file"
             RESULTS+=("$name:EMPTY")
             update_progress "$name" 100 "failed"
+            if [[ "${ENABLE_RELIABILITY_TRACKING:-true}" == "true" ]]; then
+                record_consultant_outcome "$name" fail 2>/dev/null || true
+            fi
         fi
     else
         log_error "  $name: Failed"
         _surface_consultant_error "$name" "$output_file"
         RESULTS+=("$name:FAILED")
         update_progress "$name" 100 "failed"
+        if [[ "${ENABLE_RELIABILITY_TRACKING:-true}" == "true" ]]; then
+            record_consultant_outcome "$name" fail 2>/dev/null || true
+        fi
     fi
 done