{
  "schema_version": "2.0",
  "slug": "drhalto-agentmako",
  "name": "agentmako",
  "agent_url": "https://agentmako.drhalto.com",
  "category": "Coding",
  "run_id": "run-r-publish-v2-drhalto-agentmako-2026-08-25",
  "run_at": "2026-08-25T09:00:00Z",
  "editor": "Hlido Editor",
  "editorial_method": "public-surface-tier-1+editorial-narrative-v2",
  "methodology_version": "2026.05",
  "methodology_url": "/methodology/public-surface-tier-1/",
  "score": 74,
  "tier": "STEADY",
  "laddoo_score": 74,
  "confidence": "medium",
  "hlido_opinion": {
    "headline": "A local-first MCP server that gives coding agents a live, queryable map of your repo — a broad typed-tool surface and a durable 'findings' layer, from a small but Apache-licensed project.",
    "body": "agentmako is a local-first MCP server that keeps a project's code, routes, schema, diagnostics, rules, findings and freshness state queryable for Claude Code, Codex CLI, Cursor, Cline and other MCP clients. Its distinctive idea is the Reef engine — not just an index, but a durable fact-and-finding layer that remembers what changed, what failed, what was reviewed, and whether a piece of evidence is fresh enough for an agent to trust, with a background daemon that watches edits and refreshes the map. The tool surface is unusually broad for a young project: two dozen typed tools spanning context packets, AST pattern search, route tracing, import-impact analysis, database schema and RLS inspection, a tenant-leak audit, git pre-commit checks, and project-convention extraction. Several of those (RLS inspection, tenant-leak audit) point at a genuinely useful multi-tenant-security angle most codebase tools ignore. The honesty markers are good — 'local-first, no hosted service', 'node >= 20', 'apache-2.0', typed tools — so there is no cloud dependency to trust and the license is permissive. What holds it to the lower-middle of STEADY is maturity and proof: 51 GitHub stars, no published benchmarks or independent evidence for the Reef 'freshness' claims, and a large tool surface whose reliability across real repos the public page asserts rather than demonstrates. It is a promising, well-scoped local server; it is early, and it is honest about being early.",
    "voice": "Hlido Editor",
    "as_of": "2026-08-25",
    "editor_signature_pending": true
  },
  "tier_rationale": "STEADY (74), lower-middle, because agentmako is a genuinely agent-native, local-first, Apache-licensed MCP server with a broad typed-tool surface and a thoughtful durable-findings layer (Reef), including rare multi-tenant-security tooling. It sits low in the band because it is early (51 stars), publishes no benchmarks or independent evidence for its freshness/durability claims, and its broad tool surface is asserted rather than demonstrated reliable across real projects.",
  "what_it_does_well": [
    "Local-first with no hosted service — code and queries stay on your machine",
    "Reef engine adds a durable fact-and-finding layer that survives across sessions, not just a static index",
    "Broad typed-tool surface (~two dozen) covering context, AST search, route tracing, import impact and more",
    "Rare multi-tenant-security tooling: RLS inspection and a tenant-leak audit",
    "Background daemon watches edits and tracks evidence freshness so the agent knows what to trust",
    "Apache-2.0 licensed and node >= 20 — permissive and low-dependency",
    "Works across Claude Code, Codex CLI, Cursor and Cline"
  ],
  "what_it_fails_at": [
    "Early-stage traction — around 51 GitHub stars at review",
    "No published benchmarks or independent evidence for the Reef freshness/durability claims",
    "A large tool surface whose per-tool reliability across real repos is asserted, not demonstrated",
    "Single-maintainer-scale project with no continuity or support signal published",
    "No pricing/support model because it is a free local tool — fine, but no commercial backing either"
  ],
  "best_for": [
    "Developers who want a local-first, no-cloud MCP map of their repo for a coding agent",
    "Multi-tenant SaaS teams who value the RLS-inspection and tenant-leak-audit tools",
    "Users who prefer an Apache-licensed tool they can run and inspect entirely locally"
  ],
  "not_recommended_for": [
    "Teams needing a supported, SLA-backed product with a continuity guarantee",
    "Buyers who require published benchmarks before adopting a repo-analysis tool",
    "Anyone wanting a hosted, zero-setup service rather than a local daemon"
  ],
  "red_flags": [],
  "compared_to": [
    {
      "slug": "oraios-serena",
      "verdict_diff": "Serena is a more established semantic code-navigation MCP toolkit; agentmako adds a durable findings layer (Reef) and multi-tenant-security tools but is earlier and less proven. Serena for maturity, agentmako for its findings-persistence idea and security-focused tools.",
      "preferred_for_axis": "durable-findings-layer"
    },
    {
      "slug": "neo4j-contrib-mcp-neo4j",
      "verdict_diff": "Neo4j's MCP server exposes a graph database you model; agentmako is a purpose-built local repo map with typed code tools. Neo4j's for general graph data, agentmako for out-of-the-box, local codebase context.",
      "preferred_for_axis": "local-repo-map"
    }
  ],
  "evidence_urls": [
    {
      "claim": "Local-first MCP server keeping code, routes, schema, diagnostics, rules and findings queryable for coding agents",
      "source": "https://agentmako.drhalto.com ('agentmako is a local-first MCP server that keeps code, routes, schema, diagnostics, project rules, findings, and freshness state queryable for Claude Code, Codex CLI, Cursor, Cline')",
      "tested_at": "2026-08-25",
      "verified": true
    },
    {
      "claim": "Reef engine provides a durable fact-and-finding layer with background freshness tracking",
      "source": "https://agentmako.drhalto.com ('Reef is Mako's durable fact and finding layer: it remembers what changed, what failed, what was reviewed, and which evidence is fresh enough for an agent to trust')",
      "tested_at": "2026-08-25",
      "verified": true
    },
    {
      "claim": "Apache-2.0 licensed, node >= 20, no hosted service, typed tools including RLS and tenant-leak audit",
      "source": "https://agentmako.drhalto.com ('local-first · no hosted service'; 'node >= 20'; 'apache-2.0'; tools listed include db_rls, tenant_leak_audit)",
      "tested_at": "2026-08-25",
      "verified": true
    },
    {
      "claim": "Early-stage traction of approximately 51 GitHub stars",
      "source": "https://agentmako.drhalto.com (star count '51' shown near GitHub link)",
      "tested_at": "2026-08-25",
      "verified": true
    }
  ],
  "agent_relevance": {
    "has_api": false,
    "has_cli": true,
    "has_mcp": true,
    "has_webhook": false,
    "has_sdk": false,
    "behavioral_testable": true,
    "agent_integration_path": "A local-first MCP server. An agent connects agentmako to a project ('agentmako connect .') and calls typed tools — context_packet, ast_find_pattern, route_trace, imports_impact, db_rls, tenant_leak_audit, git_precommit_check and more — to get structured repo context and durable findings. Works with Claude Code, Codex CLI, Cursor and Cline. Runs entirely locally with no API key.",
    "agent_friendly_score": 8
  },
  "checklist": [
    {
      "id": "homepage_loads",
      "pass": true,
      "required": true,
      "tested_at": "2026-08-25T09:00:00.000Z"
    },
    {
      "id": "primary_value_prop",
      "pass": true,
      "required": true,
      "evidence": "Give coding agents a live map of your repo — local-first MCP server",
      "tested_at": "2026-08-25T09:00:00.000Z"
    },
    {
      "id": "cta_present",
      "pass": true,
      "required": true,
      "evidence": "npm install -g agentmako / Read the docs",
      "tested_at": "2026-08-25T09:00:00.000Z"
    },
    {
      "id": "pricing_or_access",
      "pass": true,
      "required": false,
      "evidence": "Free, Apache-2.0, local-first with no hosted service; installed via npm",
      "tested_at": "2026-08-25T09:00:00.000Z"
    },
    {
      "id": "evidence_or_demo",
      "pass": true,
      "required": false,
      "evidence": "Documented ~two-dozen typed tools; connect/index example; docs, blog and FAQ present",
      "tested_at": "2026-08-25T09:00:00.000Z"
    }
  ],
  "summary": "A local-first MCP server that gives coding agents a live, queryable map of your repo — a broad typed-tool surface and a durable 'findings' layer, from a small but Apache-licensed project.",
  "_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": "2026-11-23",
  "review_age_days_at_publish": 0,
  "next_review_due_at": "2026-11-23",
  "attestation_url": "/data/attestations/drhalto-agentmako.json",
  "signature_pending": true,
  "source": "r-publish-editorial-v2",
  "marking_signal": {
    "checked_at": "2026-08-25",
    "source": "r-publish-editorial-enrich",
    "not_applicable": true,
    "note": "agentmako is a code-context MCP server producing structural facts and findings, not synthetic media for publication. Article-50 marking obligations do not apply. Recorded as not applicable."
  },
  "evidence_images": {
    "run_id": "run-19aa337213afcb8c-agentmako-drhalto-com",
    "base": "https://images.hlido.eu/reviews/drhalto-agentmako/run-19aa337213afcb8c-agentmako-drhalto-com",
    "files": [
      "home.png",
      "page_.png",
      "page_docs_.png",
      "page_blog_.png"
    ],
    "urls": [
      "https://images.hlido.eu/reviews/drhalto-agentmako/run-19aa337213afcb8c-agentmako-drhalto-com/home.png",
      "https://images.hlido.eu/reviews/drhalto-agentmako/run-19aa337213afcb8c-agentmako-drhalto-com/page_.png",
      "https://images.hlido.eu/reviews/drhalto-agentmako/run-19aa337213afcb8c-agentmako-drhalto-com/page_docs_.png",
      "https://images.hlido.eu/reviews/drhalto-agentmako/run-19aa337213afcb8c-agentmako-drhalto-com/page_blog_.png"
    ],
    "note": "Screenshots captured by the Hlido engine during the reviewed run, served from R2. `run_id` is the ENGINE run id — it differs from `scorecard.run_id` and is the only one these keys resolve under."
  },
  "pricing_facts": {
    "schema": "pricing-facts/1",
    "model": [
      "open-source"
    ],
    "free_tier": true,
    "pricing_disclosed": {
      "pass": true,
      "evidence": "Free, Apache-2.0, local-first with no hosted service; installed via npm",
      "tested_at": "2026-08-25"
    },
    "last_verified": "2026-08-25",
    "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": "2026-08-27"
  }
}