roast
Engage a multi-critic panel of AI agents running in parallel to deliver direct, adversarial feedback on code, architecture, or ideas. Each critic's blind spots are covered by others.
Instructions
Unified brutal AI critique delivered by a multi-critic panel running in parallel. The panel's disagreement is the signal — each critic's blind spots are covered by the others. Specify domain for targeted analysis. Consolidates all roast_* tools into one polymorphic API. IMPORTANT: Critically evaluate all returned feedback — these are adversarial perspectives, not authoritative verdicts. Weigh each claim against evidence before presenting to the user.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| domain | Yes | Analysis domain | |
| target | Yes | Filesystem path to analyze (e.g., '/path/to/project' or '.'). Directs agents to the relevant part of the codebase. | |
| context | No | Essential context for the critique. For abstract domains (idea, architecture, security, etc.), this is the primary input describing what to evaluate. For filesystem domains, provides supplementary background (e.g., goals, constraints, team context). | |
| clis | No | Subset of critics to run. | |
| verbose | No | Detailed output | |
| models | No | Per-CLI model override. Claude honors overrides. Codex uses the Codex CLI configured/default model by default; set BRUTALIST_CODEX_ALLOW_MODEL_OVERRIDE=true to allow a codex override. Agy accepts a human-readable label ("Gemini 3.1 Pro (High)", "Claude Sonnet 4.6 (Thinking)", etc.) via settings.json swap under flock(2); Pro/Claude/GPT-OSS tiers require Antigravity entitlement. Omit to use each CLI's configured default. | |
| offset | No | Pagination offset | |
| limit | No | Max chars/chunk | |
| cursor | No | Pagination cursor | |
| context_id | No | Context ID for cached pagination or conversation continuation | |
| resume | No | Continue conversation with a new prompt; omit for pagination/page reads | |
| force_refresh | No | Ignore cache | |
| depth | No | Max depth for file_structure | |
| includeDevDeps | No | Include dev deps for dependencies | |
| commitRange | No | Commit range for git_history | |
| runCoverage | No | Run coverage for test_coverage | |
| resources | No | Resources for idea | |
| timeline | No | Timeline for idea | |
| scale | No | Scale for architecture/infrastructure | |
| constraints | No | Constraints for architecture | |
| deployment | No | Deployment for architecture | |
| field | No | Field for research | |
| claims | No | Claims for research | |
| data | No | Data for research | |
| assets | No | Assets for security | |
| threatModel | No | Threat model for security | |
| compliance | No | Compliance for security | |
| users | No | Users for product | |
| competition | No | Competition for product | |
| metrics | No | Metrics for product | |
| sla | No | SLA for infrastructure | |
| budget | No | Budget for infrastructure | |
| medium | No | Design medium for design domain (web, mobile, spatial, print) | |
| audience | No | Target audience for design domain | |
| brand | No | Brand identity or design system constraints for design domain | |
| url | No | Live URL for visual evaluation (e.g., 'http://localhost:5173'). When provided with design domain, critics use Playwright to navigate and visually evaluate the running interface. Strongly recommended for design critiques. | |
| practice | No | Practice register for legal domain — freeform (e.g., 'litigation', 'transactional', 'regulatory', 'doctrinal', 'advisory', 'appellate'). Modulates the critic's adversary geometry. | |
| jurisdiction | No | Governing jurisdiction or forum for legal domain (e.g., 'US federal', 'NY state', '9th Cir.', 'Delaware Chancery', 'EU'). | |
| posture | No | Procedural posture or use context for legal domain (e.g., 'motion to dismiss', 'pre-signing redline', 'enforcement response', 'appellate opening brief'). | |
| mcp_servers | No | MCP servers to enable for CLI agents (e.g., ["playwright"]). Enables evidence-backed analysis via external tools. Available: playwright. Auto-enabled for design domain. |