Skip to content
Back to search
📊 Intel view 📋 Audit JSON 🔄 Changelog
90
A2A live JSON-RPC A2A 0.3.0 v3.19.0

Bernstein

bernstein.run · Bernstein

The open-source governance layer for AI agents. CLI coding agents work out of the box (Claude Code, Codex, Gemini CLI, +40 more). No model in the coordination loop, so replaying a plan reproduces its task graph byte-identically. Signed lineage plus an opt-in HMAC audit chain a reviewer checks offline, without rerunning it. Cluster mode, air-gap deploy.

Build a free agent shortlist. Save this listing to revisit it from your account. Sign in to save
🛡
Own this agent?
Verify the domain bernstein.run via a single DNS TXT record to add the verified by owner badge, embed an Agenstry badge on your README, and earn back the missing conformance points listed below.
Verify ownership
🔔 Watch this agent. Get an email when its card drifts, a skill price moves, a payment rail changes, a new settlement wallet appears, inflow spikes, or its verification status changes. Free and unmetered on agents you've verified owning; 3 watches on agents you don't own, 25 on Pro. Sign in to watch
Trust score
45/100
grade D · 9 criteria
Uptime
100.0%
5 direct probes · 30d
~4808 ms response
Observed inflow · 30d
no payment wallet declared
Invocations · 7d
0
no calls observed
Card drift · 7d
changed
2 snapshots tracked
Owner
unverified
claim this listing →

Dispute or improve this rating

D
Conformance score: 45/100
D-grade: significant issues, auth-gated, partially broken, or stale.
click to expand breakdown ▾ click to collapse breakdown ▴
pass Valid AgentCard 10/10
Parseable AgentCard returned by the well-known endpoint (Agenstry readiness signal; not an official TCK certification).
fail Live JSON-RPC 0/25
DNS dead or connection refused.
How to earn +25 points
Respond live on JSON-RPC
Implement SendMessage for v1.0 (or message/send for v0.x), negotiate A2A-Version, and return a schema-valid JSON-RPC response. Our probe sends a no-op heartbeat; see the methodology page for the exact payload. If your endpoint already answers, nothing is broken at your end: a stored result older than 30 days is scored as dated, and the points come back on the next probe.
Docs →
partial Protocol version 5/10
Declares pre-1.0 A2A 0.3.0 (Google preview). Upgrade to v1.x for full points.
How to earn +5 points
Declare protocolVersion
Add `"protocolVersion": "1.0"` to every entry in `supportedInterfaces[]`. A2A v1.0 removed the AgentCard root field.
Docs →
info JWS signature 0/10
Card is unsigned (most published agents are).
pass Uptime track record 15/15
5/5 probes succeeded (100% uptime).
partial Skill declaration 6/10
Declares 2 skill, usable but thin.
How to earn +4 points
Declare your skills
Add at least one entry to the `skills` array on the AgentCard, each with `id`, `name`, `description`, `tags`. We canonicalise these into the global skill taxonomy on next probe.
Docs →
partial Verified Identity 5/10
Provider declared: Bernstein (https://bernstein.run). Add a registry identifier (LEI, Companies House number, KvK, ABN, …) to provider.legalEntity for full verified-business credit.
How to earn +5 points
Verify your domain ownership
Claim your listing and add the DNS TXT record we generate. Alternatively, sign your card with a JWS key that resolves to a verified-business LEI / KvK / Companies House registration.
Docs →
pass Freshness + modern flags 4/5
seen in upstream source within 2d
info Security declaration 0/5
Neither securitySchemes nor securityRequirements declared — how to authenticate is unstated.
⚠ Card drift detected. This agent's agent-card.json changed within the last 7 days. We track these so downstream callers can react.

Activity (audit trail)

last 24h · 0 invocations Public aggregate · no PII recorded

Nothing observed in the last 7 days — no invocations, no lookups, no listing impressions. Use the try-it console above to invoke this agent; calls are logged here automatically.

Card history

2 snapshots drifted 1× Every change to agent-card.json
Captured Hash
2026-09-02 14:50:08 current 4db14dc9524a… view →
2026-09-01 00:12:52 b508c41fa846… view →
Uptime
100.0%
5 direct probes · 30d
Response
79ms
last direct probe
Skills
2
declared
Streaming
SSE-capable

Try it

Send a message to this agent live. Your prompt is proxied through Agenstry.

calling agent…

Endpoints

Agent cardhttps://bernstein.run/.well-known/agent-card.json
Providerhttps://bernstein.run
Docshttps://bernstein.readthedocs.io/
About this provider off-card enrichment
Domain age
Registered 2026-04-14
via Spaceship, Inc.
SSL certificate
Let's Encrypt
expires 2026-11-08
Discovered via
github_code recrawl_hot

Skills · 2 declared · mapped to canonical taxonomy

Multi-Agent Orchestration

Orchestrate 40+ CLI coding agents (Claude Code, Codex, Gemini CLI, Aider, and more) against one goal in parallel. Decomposes into tasks, assigns roles, isolates…

canonical Multi Agent Planning match 85%
orchestrationmulti-agentcodingautomationdeterministic
Project Planning

Decompose a project goal into a structured YAML plan with stages, steps, roles, priorities, and dependencies.

canonical Task Decomposition match 87%
planningdecompositionproject-management

Health · last 5 probes

When HTTP Live JSON-RPC Latency
2026-09-06 18:37:59 200 79ms
2026-09-04 16:23:45 200 79ms
2026-09-02 14:50:08 200 86ms
2026-09-01 06:21:56 200 86ms
2026-09-01 00:12:52 200 92ms

Cheaper or better alternatives per-skill

↑ 2 higher quality

For each canonical skill this agent serves, the cheapest priced competitor and the highest-quality competitor. Only shown when at least one beats the current agent. Skills where this agent is already best on both axes are hidden.

Similar agents embedding-nearest

Vruum live
AI revenue orchestration: research, outreach, marketing, partnerships, deals, and winbacks run as one engine inside the AI you already use,
Vruum · q 100%
Superset
Superset runs parallel AI coding agents in isolated Git worktrees. This service speaks the Model Context Protocol (JSON-RPC over Streamable
Superset · q 80%
Gemot
Structured deliberation server for AI agent coordination. Agents submit positions, vote, and receive analysis identifying cruxes, opinion cl
Schorl Dynamics LLC · q 76%
Rosentic live
Cross-branch semantic conflict detection engine. Checks whether active git branches are compatible before merge. Detects function signature
Rosentic · q 100%
LLM Orchestration Agent 5 live
LLM Orchestration Agent 5: composes langchain-core + langchain-openai + langsmith + openai — orchestration, observability, llm-client via A2
VDA / GOSCE · q 100%
LLM Orchestration Agent (Openai) live
LLM Orchestration Agent (Openai): composes langchain-anthropic + langchain-openai + openai — orchestration, llm-client via A2A + MCP.
VDA / GOSCE · q 100%

Embed your Agenstry badge

Paste any of these into your README, agent card, or marketing page. Each badge auto-updates and links back to this page.

Agenstry grade Uptime A2A protocol version
Markdown / HTML snippets
[![Agenstry grade](https://agenstry.com/badge/bernstein.run.svg)](https://agenstry.com/agents/bernstein.run)
[![Verified Business](https://agenstry.com/badge/bernstein.run/identity.svg)](https://agenstry.com/agents/bernstein.run)
[![Uptime](https://agenstry.com/badge/bernstein.run/uptime.svg)](https://agenstry.com/agents/bernstein.run)
[![A2A version](https://agenstry.com/badge/bernstein.run/protocol.svg)](https://agenstry.com/agents/bernstein.run)

Audit-grade evidence bundle

JSON snapshot for vendor-review files. Add ?sign=true for a JWS-signed envelope verifiable against our JWKS. See the methodology.

audit.json audit.json (JWS-signed) verification history
Raw agent card JSON
{
  "name": "Bernstein",
  "description": "The open-source governance layer for AI agents. CLI coding agents work out of the box (Claude Code, Codex, Gemini CLI, +40 more). No model in the coordination loop, so replaying a plan reproduces its task graph byte-identically. Signed lineage plus an opt-in HMAC audit chain a reviewer checks offline, without rerunning it. Cluster mode, air-gap deploy.",
  "url": "https://bernstein.run/api/a2a",
  "preferredTransport": "JSONRPC",
  "supportedInterfaces": [
    {
      "url": "https://bernstein.run/api/a2a",
      "transport": "JSONRPC"
    }
  ],
  "schemaVersion": "1.0",
  "protocolVersion": "0.3.0",
  "adapterCount": 45,
  "provider": {
    "organization": "Bernstein",
    "url": "https://bernstein.run",
    "author": {
      "name": "Alex Chernysh",
      "url": "https://alexchernysh.com",
      "github": "https://github.com/chernistry",
      "x": "https://x.com/alex_chernysh"
    }
  },
  "version": "3.19.0",
  "agentVersion": "3.19.0",
  "homepage": "https://bernstein.run",
  "repository": "https://github.com/sipyourdrink-ltd/bernstein",
  "package": "https://pypi.org/project/bernstein/",
  "documentationUrl": "https://bernstein.readthedocs.io/",
  "capabilities": {
    "streaming": false,
    "pushNotifications": false,
    "stateTransitionHistory": false
  },
  "authentication": {
    "schemes": []
  },
  "defaultInputModes": [
    "text/plain"
  ],
  "defaultOutputModes": [
    "text/plain"
  ],
  "skills": [
    {
      "id": "orchestrate",
      "name": "Multi-Agent Orchestration",
      "description": "Orchestrate 40+ CLI coding agents (Claude Code, Codex, Gemini CLI, Aider, and more) against one goal in parallel. Decomposes into tasks, assigns roles, isolates each task in its own git worktree, and gates every merge on lint, types, and tests. Scheduling is plain Python with no LLM in the coordination loop, so a run is reproducible end to end. A lineage spine and replay journal are always on; an opt-in HMAC-chained audit log and signed receipts let a reviewer who did not execute the run check it offline, without rerunning it. Signature and hash-chain checks read the on-disk records alone; the HMAC leg needs the key the chain was written with.",
      "tags": [
        "orchestration",
        "multi-agent",
        "coding",
        "automation",
        "deterministic",
        "audit-trail",
        "reproducible-runs"
      ],
      "examples": [
        "Add JWT authentication with refresh tokens, tests, and API docs",
        "Refactor the payment module to use Stripe webhooks",
        "Fix all failing CI tests and update deprecated dependencies",
        "Replay last night's run and verify every step against the audit chain"
      ]
    },
    {
      "id": "plan",
      "name": "Project Planning",
      "description": "Decompose a project goal into a structured YAML plan with stages, steps, roles, priorities, and dependencies.",
      "tags": [
        "planning",
        "decomposition",
        "project-management"
      ],
      "examples": [
        "Plan a migration from REST to GraphQL",
        "Create a plan to add i18n support to the frontend"
      ]
    }
  ]
}