{
  "schema_version": "2.0",
  "slug": "evermind-ai-raven",
  "name": "Raven (EverMind)",
  "agent_url": "https://raven.evermind.ai",
  "category": "Workflow & Automation",
  "run_id": "run-evermind-ai-raven-v2-rpub-2026-10-02",
  "run_at": "2026-10-02T00:30:34Z",
  "editor": "Hlido Editor",
  "editorial_method": "public-surface-tier-2+editorial-narrative-v2",
  "methodology_version": "2026.09",
  "methodology_url": "/methodology/public-surface-tier-2/",
  "score": 70,
  "tier": "STEADY",
  "laddoo_score": 70,
  "confidence": "low",
  "hlido_opinion": {
    "headline": "A multi-agent orchestration layer that sits above the coding agents you already run (Claude Code, Codex, Copilot, Qwen and more) — real GitHub traction and broad interop, wrapped in heavy 'self-evolving / RSI' language that the surface can't back up.",
    "body": "Raven, from EverMind, bills itself as 'the Harness of Harnesses' — a self-evolving multi-agent orchestration ecosystem that interprets a goal, assembles the right specialists, and keeps every handoff visible across a long list of agents you might already use (Claude Code, Codex, GitHub Copilot, Qwen Code, Grok, Kimi, and its own Raven variants). It installs on macOS/Windows via a curl one-liner and is on GitHub with a visible ~5,000-star count, which is a genuine, if early, traction signal that most entrants lack. Hlido's read: the core idea — one surface that routes a goal across whatever harnesses you have and makes the orchestration legible — is a real and useful abstraction, and the breadth of named integrations suggests actual interop work rather than vapor. What Hlido discounts is the framing. 'Self-evolving', 'Built for RSI' (recursive self-improvement) and 'Harness of Harnesses' are large claims that a landing page cannot substantiate, and Hlido did not install Raven, run an orchestration, or verify that the handoffs, presets or 'self-evolution' behave as described. Two practical notes for adopters: a 'curl | bash' installer warrants the usual scrutiny before running, and the real test of an orchestrator is reliability under multi-step handoffs, which only a hands-on run reveals. Credible and unusually well-connected for its stage; the grand self-improvement language is marketing until proven.",
    "voice": "Hlido Editor",
    "as_of": "2026-10-02",
    "editor_signature_pending": true
  },
  "tier_rationale": "STEADY (70) rests on two real signals: broad, concretely-named interop across the major coding agents, and visible GitHub traction (~5k stars) that most new entrants do not have. It is held at the STEADY floor rather than higher because the flagship 'self-evolving / RSI / Harness of Harnesses' framing is unverified, Hlido did not run it, and confidence is low on a single-surface read — orchestration reliability can only be judged hands-on.",
  "what_it_does_well": [
    "Clear, useful abstraction: one surface that routes a goal across the harnesses you already run and keeps handoffs visible",
    "Broad, concretely-named interop (Claude Code, Codex, GitHub Copilot, Qwen Code, Grok, Kimi and its own Raven variants)",
    "Visible GitHub traction (~5,000 stars) — a real early-adoption signal",
    "Cross-platform install (macOS/Windows) with presets to lower the starting effort"
  ],
  "what_it_fails_at": [
    "'Self-evolving' / 'Built for RSI' / 'Harness of Harnesses' are large claims unsupported by the public surface",
    "Not installed or exercised by Hlido — orchestration reliability, handoff quality and 'self-evolution' are unverified",
    "'curl | bash' install path warrants security scrutiny before running",
    "No pricing/limits or data-handling detail captured from the surface"
  ],
  "best_for": [
    "Developers already juggling multiple coding agents who want a single orchestration surface over them",
    "Teams that value visible handoffs and presets when composing multi-agent workflows",
    "Early adopters comfortable evaluating a fast-moving open project hands-on"
  ],
  "not_recommended_for": [
    "Teams needing a stable, supported orchestrator with documented reliability guarantees",
    "Security-conservative environments uneasy with a curl-pipe installer",
    "Buyers who would take 'self-evolving / RSI' claims at face value without testing"
  ],
  "red_flags": [
    "Markets 'self-evolving' and 'Built for RSI' (recursive self-improvement) — strong claims not substantiated on the public surface; verify hands-on before relying on them",
    "Primary install is a 'curl | bash' one-liner — review the script before executing"
  ],
  "compared_to": [],
  "evidence_urls": [
    {
      "claim": "Orchestrates/routes goals across multiple existing agents (Claude Code, Codex, Copilot, Qwen, etc.)",
      "source": "https://raven.evermind.ai/ (homepage orchestration section, captured)",
      "tested_at": "2026-10-02",
      "verified": true
    },
    {
      "claim": "Installable on macOS/Windows; public GitHub repo with ~5,000 stars",
      "source": "https://raven.evermind.ai/ (install one-liner + star count, captured)",
      "tested_at": "2026-10-02",
      "verified": true
    },
    {
      "claim": "Self-evolving / built for recursive self-improvement (RSI)",
      "source": "https://raven.evermind.ai/ (headline claim; NOT tested or verified)",
      "tested_at": "2026-10-02",
      "verified": false
    }
  ],
  "agent_relevance": {
    "has_api": false,
    "has_cli": true,
    "has_mcp": false,
    "has_webhook": false,
    "has_sdk": false,
    "behavioral_testable": true,
    "agent_integration_path": "Raven is itself an agent-orchestration layer: installed via CLI (curl one-liner), it drives and coordinates other coding agents you already run. Open on GitHub and installable, so it is behaviorally testable by running it; no separate API/MCP surface is advertised for an outside agent to drive Raven itself.",
    "agent_friendly_score": 7
  },
  "checklist": [
    {
      "id": "homepage_loads",
      "pass": true,
      "required": true,
      "tested_at": "2026-10-02T00:30:34Z"
    },
    {
      "id": "primary_value_prop",
      "pass": true,
      "required": true,
      "evidence": "'Every agent, one surface' — multi-agent orchestration",
      "tested_at": "2026-10-02T00:30:34Z"
    },
    {
      "id": "cta_present",
      "pass": true,
      "required": true,
      "evidence": "Install one-liner + 'View On GitHub'",
      "tested_at": "2026-10-02T00:30:34Z"
    },
    {
      "id": "pricing_or_access",
      "pass": true,
      "required": false,
      "evidence": "Free install via curl + public GitHub; no paid tiers captured",
      "tested_at": "2026-10-02T00:30:34Z"
    },
    {
      "id": "evidence_or_demo",
      "pass": true,
      "required": false,
      "evidence": "'See in run' section + public repo; not exercised by Hlido",
      "tested_at": "2026-10-02T00:30:34Z"
    }
  ],
  "summary": "A multi-agent orchestration layer that sits above the coding agents you already run (Claude Code, Codex, Copilot, Qwen and more) — real GitHub traction and broad interop, wrapped in heavy 'self-evolving / RSI' language that the surface can't back up.",
  "_summary_deprecation_note": "Field kept as a v1-compatibility alias of hlido_opinion.headline. New consumers should read hlido_opinion.{headline,body,voice,as_of}.",
  "staleness_after": "2027-01-02",
  "review_age_days_at_publish": 0,
  "next_review_due_at": "2027-01-02",
  "attestation_url": "/data/attestations/evermind-ai-raven.json",
  "signature_pending": true,
  "source": "hlido-editor-v2",
  "marking_signal": {
    "not_applicable": true,
    "checked_at": "2026-10-02",
    "source": "r-publish-editorial-2026-10-02"
  },
  "pricing_facts": {
    "schema": "pricing-facts/1",
    "pricing_disclosed": {
      "pass": true,
      "evidence": "Free install via curl + public GitHub; no paid tiers captured",
      "tested_at": "2026-10-02"
    },
    "last_verified": "2026-10-02",
    "basis": "Derived from Hlido-held evidence only (engine checklist + editorial text); quotes are verbatim from the scorecard; not vendor-supplied; re-derived daily. Verify current prices on the vendor's pricing page.",
    "derived_at": "2026-10-03"
  }
}