> The Markdown version of [RACI](raci.html) on the aayudha docs site. Every page is listed in [llms.txt](llms.txt).

<!-- Code generated by go generate (internal/verbs/gen); DO NOT EDIT. -->

# RACI

Generated from `internal/verbs/registry.go` and `internal/verbs/raci.go` by `go generate ./...` (WP5.1).
Every verb has exactly one Accountable seat (A). The Responsible party (R) is always that seat's agent twin working in
the verb's model lane, with the seat's paired human when it has one; the seat that owns the verb's thresholds or policy
may differ (the AI Native Engineer executes `verify` against floors the Quality Architect owns in
`CONSTRAINTS.md`; the Engineering Manager owns the `inspect` policy that the review lane executes).
Consulted seats (C) are asked before the verb acts and Informed seats (I) are told after. Every layer has exactly one
owner seat (LAYERS.md). The registry lint (`verbs.LintRACI`) fails generation otherwise. Each seat runs
solo (agent only) or paired (agent twin plus a human): `[seats.<name>]`, set by `aayu seat add|remove`.
Charters: `seats/<seat>/CHARTER.md`.

## Verbs

| Verb | A | R (executes) | Owns thresholds / policy | C | I | Layers | Identity | Model calling | MCP |
|---|---|---|---|---|---|---|---|---|---|
| `grill` | Product Strategist | Product Strategist twin, spec lane | Product Strategist | Code Architect, UX Researcher & Evolutionist | Engineering Manager | all | bot | host | read |
| `slice` | Product Strategist | Product Strategist twin, plan lane | Product Strategist | Code Architect, Engineering Manager | AI Native Engineer | all | bot | host | read |
| `plan` | Code Architect | Code Architect twin, plan lane | Code Architect | Engineering Manager, Quality Architect, Data & AI Architect, Production Engineer | AI Native Engineer | all | bot | host | read |
| `decide` | Founder | Founder twin, spec lane | Founder | Product Strategist, Code Architect | Engineering Manager, Auditor | all | human | host | read |
| `counsel` | Founder | Founder twin, spec lane | Founder | Product Strategist, Code Architect, Engineering Manager | Auditor | all | bot | host | read |
| `triage` | Product Strategist | Product Strategist twin, chore lane | Product Strategist | — | Engineering Manager | none | bot | hybrid (the host's model passes --as; never in product CI) | write |
| `brain` | Product Strategist | Product Strategist twin, chore lane | Product Strategist | Founder, Revenue Architect | Marketer & Documenter | none | bot | none | read |
| `build` | AI Native Engineer | AI Native Engineer twin, build lane | AI Native Engineer | Code Architect, Production Engineer | Engineering Manager | all | bot | host | write |
| `verify` | AI Native Engineer | AI Native Engineer twin, chore lane | Quality Architect | — | Engineering Manager | all | bot | none | write |
| `inspect` | Engineering Manager | Engineering Manager twin, review lane | Engineering Manager | Quality Architect, Code Architect | AI Native Engineer | all | bot | direct | write |
| `ship` | Engineering Manager | Engineering Manager twin, chore lane | Engineering Manager | Quality Architect | Product Strategist, Marketer & Documenter | all | human | none | never |
| `release` | DevOps & DevEx Tinkerer | DevOps & DevEx Tinkerer twin, chore lane | DevOps & DevEx Tinkerer | Quality Architect, Engineering Manager, Production Engineer | Marketer & Documenter, Founder | L09 | human | none | write |
| `proof` | Quality Architect | Quality Architect twin, chore lane | Quality Architect | AI Native Engineer | Auditor | all | bot | none | read |
| `ready` | Product Strategist | Product Strategist twin, chore lane | Product Strategist | Code Architect, Quality Architect | Engineering Manager | all | bot | none | read |
| `research` | UX Researcher & Evolutionist | UX Researcher & Evolutionist twin, spec lane | UX Researcher & Evolutionist | Product Strategist | Marketer & Documenter | L01 | bot | host | read |
| `wireframe` | UX Researcher & Evolutionist | UX Researcher & Evolutionist twin, build lane | UX Researcher & Evolutionist | Product Strategist | Engineering Manager | L01 | bot | host | write |
| `mockup` | UX Researcher & Evolutionist | UX Researcher & Evolutionist twin, build lane | UX Researcher & Evolutionist | Product Strategist | Engineering Manager | L01 | bot | host | write |
| `prototype` | UX Researcher & Evolutionist | UX Researcher & Evolutionist twin, build lane | UX Researcher & Evolutionist | Code Architect | Engineering Manager | L01, L02 | bot | host | write |
| `review` | UX Researcher & Evolutionist | UX Researcher & Evolutionist twin, ui-review lane | UX Researcher & Evolutionist | Quality Architect | Engineering Manager | L01 | bot | hybrid (lane A direct) | write |
| `polish` | UX Researcher & Evolutionist | UX Researcher & Evolutionist twin, build lane | UX Researcher & Evolutionist | — | Engineering Manager | L01 | bot | host | write |
| `baseline` | UX Researcher & Evolutionist | UX Researcher & Evolutionist twin, chore lane | UX Researcher & Evolutionist | Quality Architect | Engineering Manager | L01 | bot | none | write |
| `experiment` | Revenue Architect | Revenue Architect twin, plan lane | Revenue Architect | Product Strategist, UX Researcher & Evolutionist, Data & AI Architect | Founder | L03, L08 | bot | host | write |
| `readout` | Revenue Architect | Revenue Architect twin, spec lane | Revenue Architect | Product Strategist | Founder, Marketer & Documenter | L08 | bot | host | write |
| `respond` | Production Engineer | Production Engineer twin, build lane | Production Engineer | AI Native Engineer, DevOps & DevEx Tinkerer | Engineering Manager, Founder | L07, L09 | human | host | never |
| `status` | Engineering Manager | Engineering Manager twin, chore lane | Engineering Manager | — | Founder | all | bot | none | read |
| `audit` | Auditor | Auditor twin, review lane | Auditor | Quality Architect | Founder | all | bot | none | write |
| `retro` | Engineering Manager | Engineering Manager twin, spec lane | Engineering Manager | Auditor, AI Native Engineer | Founder | all | bot | host | write |
| `measure` | Data & AI Architect | Data & AI Architect twin, spec lane | Revenue Architect | Product Strategist | Founder | L08 | bot | none | write |
| `evolve` | AI Native Engineer | AI Native Engineer twin, build lane | AI Native Engineer | Quality Architect, Data & AI Architect | Engineering Manager | L05 | bot | host | write |
| `brief` | Mentalist | Mentalist twin, spec lane | Mentalist | Product Strategist | Marketer & Documenter | all | bot | hybrid (adversarial pass direct) | read |
| `tell` | Marketer & Documenter | Marketer & Documenter twin, spec lane | Marketer & Documenter | Mentalist, Product Strategist | Founder | L01 | human | host (publish requires the human identity) | write |
| `init` | Founder | Founder twin, chore lane | Founder | DevOps & DevEx Tinkerer | Engineering Manager | L09 | human | none | never |
| `install` | DevOps & DevEx Tinkerer | DevOps & DevEx Tinkerer twin, chore lane | DevOps & DevEx Tinkerer | — | Engineering Manager | L09 | human | none | never |
| `doctor` | DevOps & DevEx Tinkerer | DevOps & DevEx Tinkerer twin, chore lane | DevOps & DevEx Tinkerer | — | Engineering Manager | L09 | human | none | read |
| `chart` | Data & AI Architect | Data & AI Architect twin, spec lane | Data & AI Architect | Code Architect, Revenue Architect | AI Native Engineer, Quality Architect | L05, L07, L08 | bot | host | read |
| `survey` | Code Architect | Code Architect twin, plan lane | Code Architect | Product Strategist, Quality Architect | Engineering Manager | all | bot | host | read |
| `investigate` | Production Engineer | Production Engineer twin, spec lane | Production Engineer | AI Native Engineer | Engineering Manager | all | bot | host | read |
| `slo` | Production Engineer | Production Engineer twin, spec lane | Production Engineer | Product Strategist, Quality Architect | Engineering Manager, DevOps & DevEx Tinkerer | all | bot | host | read |
| `runbook` | Production Engineer | Production Engineer twin, spec lane | Production Engineer | AI Native Engineer, DevOps & DevEx Tinkerer | Engineering Manager | all | bot | host | read |
| `scope` | Product Strategist | Product Strategist twin, plan lane | Product Strategist | Engineering Manager, Founder | DevOps & DevEx Tinkerer, Marketer & Documenter | all | bot | host | write |

Model calling: `none` is deterministic; `host` means the host's model reasons inside the skill;
`direct` means the binary calls a model itself; `hybrid` is host reasoning plus a direct pass
(ground rule 5). MCP: `read` is exposed by default, `write` only with `[mcp] expose_writes = true`,
`never` is human-only or admin (WP3.3).

## Plumbing commands

| Command | Accountable | Identity |
|---|---|---|
| `config` | DevOps & DevEx Tinkerer | human |
| `auth` | DevOps & DevEx Tinkerer | human |
| `actions` | DevOps & DevEx Tinkerer | human |
| `models` | DevOps & DevEx Tinkerer | human |
| `mcp` | DevOps & DevEx Tinkerer | bot |
| `self-update` | DevOps & DevEx Tinkerer | human |
| `upgrade` | DevOps & DevEx Tinkerer | human |
| `version` | DevOps & DevEx Tinkerer | human |
| `halt` | Founder | human |
| `resume` | Founder | human |
| `gc` | DevOps & DevEx Tinkerer | human |
| `uninit` | DevOps & DevEx Tinkerer | human |
| `hook` | DevOps & DevEx Tinkerer | bot |
| `seat` | Founder | human |
| `lfs-agent` | DevOps & DevEx Tinkerer | human |
| `next` | Engineering Manager | human |
| `tour` | Founder | human |
| `inbox` | Founder | human |
| `completion` | DevOps & DevEx Tinkerer | human |

## Seats

| Seat | Slug |
|---|---|
| Product Strategist | `product-strategist` |
| Code Architect | `code-architect` |
| Data & AI Architect | `data-ai-architect` |
| Engineering Manager | `engineering-manager` |
| AI Native Engineer | `ai-native-engineer` |
| Production Engineer | `production-engineer` |
| UX Researcher & Evolutionist | `ux-researcher-evolutionist` |
| DevOps & DevEx Tinkerer | `devops-devex-tinkerer` |
| Quality Architect | `quality-architect` |
| Mentalist | `mentalist` |
| Auditor | `auditor` |
| Marketer & Documenter | `marketer-documenter` |
| Revenue Architect | `revenue-architect` |
| Founder | `founder` |

## Model lanes

Settled defaults per host. Configuration (`[models.<lane>.<host>]`, WP1.2) overrides them; the model gateway
resolves them (WP1.6). Model IDs are as the AI gateway lists them; the family belongs to the model, not the endpoint.
Cursor cannot run the openai family, so a GPT model configured for Cursor falls back to the lane's Cursor default.
Cursor's defaults follow `[hosts.cursor] plan` (the review-ladder decision of 2026-10-17): `x`, the default, for a plan that includes only
Cursor's own models, or `legacy` for one that still includes Claude and Gemini.
When the binary calls a model itself and the host's default is a host-managed mode, it uses the Claude Code default.

The reviewer model family must differ from the builder model family on every story (reviewer.family != builder.family).

| Lane | Host | Default | Model ID | Family | Escalation / alternative |
|---|---|---|---|---|---|
| `build` | Amp | Amp medium | host-managed | host-managed | escalation: Amp high |
| `build` | Cursor (x plan) | Grok 4.7 | `grok-4.7` | xai | — |
| `build` | Cursor (legacy plan) | Claude Opus 5.5 | `claude-opus-5-5` | anthropic | or Claude Sonnet 5.5 (claude-sonnet-5-5) |
| `build` | Claude Code | Claude Opus 5.5 | `claude-opus-5-5` | anthropic | or Claude Sonnet 5.5 (claude-sonnet-5-5) |
| `build` | OpenAI Codex | GPT-6.1 Sol | `gpt-6.1-sol` | openai | escalation: GPT-6 Astra at high reasoning effort (`gpt-6-astra`) |
| `build` | GitHub Copilot | Claude Sonnet 5.5 | `claude-sonnet-5.5` | anthropic | escalation: Claude Opus 5.5 (`claude-opus-5.5`) |
| `spec` | Amp | Amp medium | host-managed | host-managed | escalation: Amp high |
| `spec` | Cursor (x plan) | Grok 4.7 | `grok-4.7` | xai | — |
| `spec` | Cursor (legacy plan) | Claude Opus 5.5 | `claude-opus-5-5` | anthropic | or Claude Sonnet 5.5 (claude-sonnet-5-5) |
| `spec` | Claude Code | Claude Opus 5.5 | `claude-opus-5-5` | anthropic | or Claude Sonnet 5.5 (claude-sonnet-5-5) |
| `spec` | OpenAI Codex | GPT-6.1 Sol | `gpt-6.1-sol` | openai | escalation: GPT-6 Astra at high reasoning effort (`gpt-6-astra`) |
| `spec` | GitHub Copilot | Claude Sonnet 5.5 | `claude-sonnet-5.5` | anthropic | escalation: Claude Opus 5.5 (`claude-opus-5.5`) |
| `plan` | Amp | Amp medium | host-managed | host-managed | escalation: Amp high |
| `plan` | Cursor (x plan) | Grok 4.7 | `grok-4.7` | xai | — |
| `plan` | Cursor (legacy plan) | Claude Opus 5.5 | `claude-opus-5-5` | anthropic | or Claude Sonnet 5.5 (claude-sonnet-5-5) |
| `plan` | Claude Code | Claude Opus 5.5 | `claude-opus-5-5` | anthropic | or Claude Sonnet 5.5 (claude-sonnet-5-5) |
| `plan` | OpenAI Codex | GPT-6.1 Sol | `gpt-6.1-sol` | openai | escalation: GPT-6 Astra at high reasoning effort (`gpt-6-astra`) |
| `plan` | GitHub Copilot | Claude Sonnet 5.5 | `claude-sonnet-5.5` | anthropic | escalation: Claude Opus 5.5 (`claude-opus-5.5`) |
| `review` (adversarial) | Amp | GPT-6.1 Sol | `gpt-6.1-sol` | openai | escalation: GPT-6 Astra at high reasoning effort for heavy reviews (`gpt-6-astra`) |
| `review` (adversarial) | Cursor (x plan) | Grok 4.6 | `grok-4.6` | xai | — |
| `review` (adversarial) | Cursor (legacy plan) | Grok 4.7 | `grok-4.7` | xai | — |
| `review` (adversarial) | Claude Code | GPT-6.1 Sol | `gpt-6.1-sol` | openai | escalation: GPT-6 Astra at high reasoning effort for heavy reviews (`gpt-6-astra`) |
| `review` (adversarial) | OpenAI Codex | Claude Opus 5.5 | `claude-opus-5-5` | anthropic | or Grok 4.7 (grok-4.7) |
| `review` (adversarial) | GitHub Copilot | GPT-6.1 Sol | `gpt-6.1-sol` | openai | or Claude Opus 5.5 (claude-opus-5.5) when the build has no anthropic response, Gemini 4 Argon (gemini-4-argon) when it used both anthropic and openai |
| `ui-review` (adversarial) | Amp | Amp high | host-managed | host-managed | — |
| `ui-review` (adversarial) | Cursor (x plan) | Grok 4.6 | `grok-4.6` | xai | — |
| `ui-review` (adversarial) | Cursor (legacy plan) | Gemini 3.8 Flash (high) | `gemini-3.8-flash` | google | or Fable 5.1 (claude-fable-5-1; anthropic family, so only when the builder is not anthropic) |
| `ui-review` (adversarial) | Claude Code | Gemini 3.8 Flash (high) | `gemini-3.8-flash` | google | or Fable 5.1 (claude-fable-5-1; anthropic family, so only when the builder is not anthropic) |
| `ui-review` (adversarial) | OpenAI Codex | Gemini 3.8 Flash (high) | `gemini-3.8-flash` | google | or Fable 5.1 (claude-fable-5-1; anthropic family, so only when the builder is not anthropic) |
| `ui-review` (adversarial) | GitHub Copilot | Gemini 4 Argon (high) | `gemini-4-argon` | google | or Gemini 3.8 Flash (high) (gemini-3.8-flash) through the AI gateway when copilot is not installed or signed in, a call fails, or [models.ui-review.copilot] model = ""; Argon costs about 0.043 USD per lane A call (Copilot AI credits: input x 2 USD/M + visible output x 10 USD/M; reasoning not billed) |
| `chore` | Amp | Amp low | host-managed | host-managed | — |
| `chore` | Cursor (x plan) | Composer 2.5 | `composer-2.5` | unknown | note: family unknown: Cursor does not publish Composer's base model |
| `chore` | Cursor (legacy plan) | Haiku 5.5 (100k context) | `claude-haiku-5-5` | anthropic | — |
| `chore` | Claude Code | Haiku 5.5 (100k context) | `claude-haiku-5-5` | anthropic | — |
| `chore` | OpenAI Codex | Haiku 5.5 (100k context) | `claude-haiku-5-5` | anthropic | — |
| `chore` | GitHub Copilot | Copilot Auto (Efficiency) | `auto:efficiency` | host-managed | — |