{
 "kind": "mcp",
 "slug": "dev-emisar-emisar",
 "path": "mcp/servers/dev-emisar-emisar/",
 "name": "emisar",
 "maker": "andrewdryga",
 "density": "full",
 "crumbs": [
  [
   "MCP servers",
   "mcp/"
  ],
  [
   "Other",
   "mcp/other/"
  ]
 ],
 "lead": "let your AI work on servers with approvals.",
 "chips": [
  {
   "label": "Other",
   "path": "mcp/other/"
  },
  {
   "label": "Official"
  },
  {
   "label": "Remote"
  }
 ],
 "primary": {
  "label": "Copy connect command",
  "copy": "claude mcp add --transport http emisar https://emisar.dev/api/mcp/rpc"
 },
 "links": [
  {
   "label": "GitHub",
   "href": "https://github.com/andrewdryga/emisar"
  },
  {
   "label": "Website",
   "href": "https://emisar.dev"
  }
 ],
 "track": "New versions and outages of the remote endpoint on your dashboard",
 "verdicts": [
  {
   "key": "official",
   "label": "Made by",
   "value": "The service itself",
   "status": "yes",
   "sub": "Official server",
   "quote": "portal/   Elixir/Phoenix control plane, operator console, website, and MCP API",
   "note": "The repo holds emisar's own control plane, bridge and runner; its README sends users to sign up at emisar.dev, the site links back to this repo, and the registry name dev.emisar sits on that domain.",
   "src": "https://github.com/AndrewDryga/emisar/blob/31f4c55b3f7513621e09e3f4468ff29dfdc6e123/README.md"
  },
  {
   "key": "secrets",
   "label": "Keys",
   "value": "1 secret needed",
   "status": "info",
   "sub": "EMISAR_API_KEY",
   "quote": "Claude.ai, ChatGPT, and other remote OAuth clients should connect directly.",
   "note": "Claude.ai and ChatGPT connect to the remote address with OAuth sign-in; Claude Code, Codex, Cursor local mode and other local clients use the emisar-mcp bridge, whose connect command writes their configs. The console's AI agents page generates the exact per-client lines; the README gives no Claude Code or Codex command. The cursor line is dist/cursor-plugin/mcp.json.",
   "src": "https://github.com/AndrewDryga/emisar/blob/31f4c55b3f7513621e09e3f4468ff29dfdc6e123/mcp/README.md"
  },
  {
   "key": "where",
   "label": "Runs",
   "value": "Remote",
   "status": "yes",
   "sub": "Nothing runs on your machine",
   "note": "Endpoint: https://emisar.dev/api/mcp/rpc. Checked: yes: an unauthenticated GET to https://emisar.dev/api/mcp/rpc on 2026-11-10 answered 405 (the endpoint takes POST; real calls need sign-in).",
   "src": "https://emisar.dev/api/mcp/rpc"
  },
  {
   "key": "license",
   "label": "Commercial use",
   "value": "With conditions",
   "status": "cond",
   "sub": "Custom license license",
   "quote": "Other production use requires a commercial license.",
   "note": "Use the hosted service under its plans (free for 3 runners and 1 user); running the control-plane code yourself in production needs a commercial license.",
   "src": "https://github.com/AndrewDryga/emisar/blob/31f4c55b3f7513621e09e3f4468ff29dfdc6e123/README.md"
  }
 ],
 "vfoot": "Checked against its repository at one commit, on Nov 10, 2026.",
 "blocks": [
  {
   "id": "overview",
   "nav": "Overview",
   "h2": "What emisar does",
   "parts": [
    {
     "t": "prose",
     "text": "emisar lets an AI assistant check and fix your servers without handing it full remote access. It can only pick from a list of prepared actions, like reading logs or restarting a service. Your rules decide what runs, what waits for a person to approve, and what is refused, and every step is logged. It needs an emisar account and a small program on each server."
    },
    {
     "t": "h3",
     "text": "What people use it for"
    },
    {
     "t": "uses",
     "items": [
      {
       "title": "Look into an incident on live servers",
       "text": "When something breaks, your assistant reads logs, service status and disk use across your servers and comes back with evidence and a likely cause.",
       "idea": false
      },
      {
       "title": "Restarts that wait for your yes",
       "text": "Routine checks run on their own, while restarts and deployments pause until a person approves them in the emisar console.",
       "idea": false
      },
      {
       "title": "Find and run a prepared action",
       "text": "The assistant searches the action list by task, reads the exact inputs one action takes, then runs it on the servers you name.",
       "idea": false
      },
      {
       "title": "Check what your own deploy did",
       "text": "After a release misbehaves, the assistant looks at the running result, finds its own mistake and proposes a fix for you to review.",
       "idea": false
      },
      {
       "title": "Turn a fix into a reviewed runbook",
       "text": "The assistant writes a repeatable list of steps as a draft; a person reviews and publishes it before anyone can run it.",
       "idea": false
      },
      {
       "title": "See what the assistant did last night",
       "text": "Review every run the assistant started, with who asked, the reason, the policy decision and the result.",
       "idea": false
      }
     ]
    }
   ]
  },
  {
   "id": "popularity",
   "nav": "Popularity",
   "h2": "emisar popularity",
   "parts": [
    {
     "t": "stats",
     "items": [
      {
       "n": "449",
       "l": "GitHub stars, andrewdryga/emisar"
      },
      {
       "n": "0",
       "l": "public agent projects connect it"
      }
     ]
    }
   ]
  },
  {
   "id": "tools",
   "nav": "Tools",
   "h2": "Tools it gives your agent (14)",
   "parts": [
    {
     "t": "tools",
     "items": [
      {
       "name": "list_packs",
       "text": "Lists the trusted action packs your servers offer and any deployment problems. Read-only."
      },
      {
       "name": "list_runners",
       "text": "Shows your servers (runners), whether they are connected, and which packs they run. Read-only."
      },
      {
       "name": "find_actions",
       "text": "Searches the actions you may run by task, across all servers in one search. Read-only."
      },
      {
       "name": "get_action",
       "text": "Returns one action's exact inputs, risk level, side effects and the servers it can run on. Read-only."
      },
      {
       "name": "run_action",
       "text": "Runs one action on named servers after policy and any human approval. Can change or delete things: marked destructive."
      },
      {
       "name": "get_operation",
       "text": "Looks up a change already requested, after a lost reply, without repeating it. Read-only."
      },
      {
       "name": "wait_for_run",
       "text": "Waits for a run to change or finish and streams its output. Read-only."
      },
      {
       "name": "cancel_run",
       "text": "Cancels one of your own runs while it still waits in the queue or for approval."
      },
      {
       "name": "recent_runs",
       "text": "Pages through the history of runs started through this connection. Read-only."
      },
      {
       "name": "list_runbooks",
       "text": "Lists runbooks (saved step lists) with their live version and any unpublished change. Read-only."
      },
      {
       "name": "get_runbook",
       "text": "Reads one runbook, its live version or its unpublished change. Read-only."
      },
      {
       "name": "execute_runbook",
       "text": "Runs a runbook's live version through policy and approval. May contain destructive steps."
      },
      {
       "name": "create_runbook_draft",
       "text": "Writes a new runbook as a draft for a person to review; nothing runs or gets published."
      },
      {
       "name": "update_runbook_draft",
       "text": "Replaces a runbook's single unpublished draft; publishing stays with a person."
      }
     ]
    }
   ]
  },
  {
   "id": "data",
   "nav": "Data and license",
   "h2": "What it sends and its license",
   "parts": [
    {
     "t": "rows",
     "rows": [
      {
       "l": "Sends",
       "v": "every action request with its exact arguments, policy decisions, approvals and the redacted output",
       "sub": "To emisar control plane on Google Cloud in the United States; every tool call that runs or records an action. Opt out: none; mark sensitive action fields so they are masked in views (they stay in storage)."
      },
      {
       "l": "Sends",
       "v": "command output from your servers, filtered on the server before it leaves",
       "sub": "To emisar run history; every run. Opt out: per-action redaction rules."
      },
      {
       "l": "Sends",
       "v": "product usage events (not your servers' data or action contents)",
       "sub": "To Mixpanel; while you use emisar. Opt out: none stated."
      }
     ]
    },
    {
     "t": "prose",
     "text": "The server program, bridge and action packs are open source under Apache 2.0. The control plane is under the Business Source License, and the hosted service is free for 3 servers and paid per server above that."
    },
    {
     "t": "rows",
     "rows": [
      {
       "l": "In a paid app",
       "v": "With conditions",
       "status": "cond",
       "sub": "Use the hosted service under its plans (free for 3 runners and 1 user); running the control-plane code yourself in production needs a commercial license.",
       "quote": "Other production use requires a commercial license.",
       "src": "https://github.com/AndrewDryga/emisar/blob/31f4c55b3f7513621e09e3f4468ff29dfdc6e123/README.md"
      },
      {
       "l": "Pass on its files",
       "v": "With conditions",
       "status": "cond",
       "sub": "The runner, bridge, packs, skills and Cursor plugin are Apache 2.0; the rest of the code is Business Source License 1.1 until its change date.",
       "quote": "`runner/`, `mcp/`, `packs/`, `skills/`, and `dist/cursor-plugin/` are open source under the [Apache License 2.0](runner/LICENSE)",
       "src": "https://github.com/AndrewDryga/emisar/blob/31f4c55b3f7513621e09e3f4468ff29dfdc6e123/README.md"
      },
      {
       "l": "client-work",
       "v": "With conditions",
       "status": "cond",
       "sub": "Fine on the hosted service under a plan that fits the client's server count; each person's key acts with their own access.",
       "quote": "Free for 3 runners",
       "src": "https://emisar.dev/"
      }
     ]
    }
   ]
  },
  {
   "id": "alive",
   "nav": null,
   "spy_as": "data",
   "h2": "Is it alive",
   "parts": [
    {
     "t": "rows",
     "rows": [
      {
       "l": "Latest version",
       "v": "0.60.0 in the registry, updated Nov 18, 2026"
      },
      {
       "l": "Last release",
       "v": "0.58.0 in the MCP Registry (published 2026-11-08); code commit 31f4c55 of 2026-11-09"
      },
      {
       "l": "Remote endpoint",
       "v": "yes: an unauthenticated GET to https://emisar.dev/api/mcp/rpc on 2026-11-10 answered 405 (the endpoint takes POST; real calls need sign-in)",
       "status": "yes"
      }
     ]
    }
   ]
  },
  {
   "id": "alternatives",
   "nav": "Alternatives",
   "h2": "Alternatives",
   "parts": [
    {
     "t": "alts",
     "items": [
      {
       "name": "emisar",
       "me": true,
       "facts": [
        "449 stars",
        "Remote"
       ]
      },
      {
       "name": "ansible-core",
       "why": "Pick it if you want repeatable server configuration written as reviewed playbooks",
       "path": null,
       "facts": [
        "Local"
       ]
      },
      {
       "name": "mcp-server-commands",
       "why": "Pick it if the agent only needs to run commands on your own disposable machine",
       "path": "mcp/servers/io-github-bytedance-mcp-server-commands/",
       "facts": [
        "39.6K stars",
        "Local",
        "in 1 projects"
       ]
      },
      {
       "name": "io-github-coder-coder",
       "why": "Pick it if the agent should manage cloud development workspaces",
       "path": null,
       "facts": [
        "Local"
       ]
      }
     ],
     "more": "All other servers",
     "more_path": "mcp/other/"
    }
   ]
  },
  {
   "id": "start",
   "nav": "Get started",
   "h2": "Get started",
   "parts": [
    {
     "t": "prompt",
     "text": "Use the emisar server to find out why the api service on web-1 is failing; use read-only actions first, give a reason on every call, and ask me before any action that needs approval."
    },
    {
     "t": "h3",
     "text": "Connect it"
    },
    {
     "t": "tabs",
     "label": "Connect by client",
     "items": [
      {
       "tab": "Claude Code",
       "code": "claude mcp add --transport http emisar https://emisar.dev/api/mcp/rpc"
      },
      {
       "tab": "Cursor",
       "code": "{\"mcpServers\": {\"emisar\": {\"type\": \"http\", \"url\": \"https://emisar.dev/api/mcp/rpc\"}}}"
      },
      {
       "tab": "Any client (JSON)",
       "code": "{\n  \"mcpServers\": {\n    \"emisar\": {\n      \"url\": \"https://emisar.dev/api/mcp/rpc\"\n    }\n  }\n}"
      }
     ]
    },
    {
     "t": "small",
     "text": "Claude.ai and ChatGPT connect to the remote address with OAuth sign-in; Claude Code, Codex, Cursor local mode and other local clients use the emisar-mcp bridge, whose connect command writes their configs. The console's AI agents page generates the exact per-client lines; the README gives no Claude Code or Codex command. The cursor line is dist/cursor-plugin/mcp.json."
    },
    {
     "t": "h3",
     "text": "Mistakes to avoid"
    },
    {
     "t": "mistakes",
     "items": [
      {
       "dont": "Calling run_action again after its reply was lost",
       "fix": "Recover it with get_operation and the same key; never repeat the change",
       "why": "A lost reply does not undo work the control plane already accepted."
      },
      {
       "dont": "Calling run_action straight from a find_actions result",
       "fix": "Call get_action first for the exact inputs, risk and side effects, then run_action",
       "why": "The selected action may be destructive and its inputs are strict."
      },
      {
       "dont": "Running find_actions once per server",
       "fix": "One find_actions query covers every server in scope",
       "why": "Repeated searches only waste calls."
      },
      {
       "dont": "Committing .mcp.json with EMISAR_API_KEY to the project",
       "fix": "Keep the key in user-level or secret config; one key per client",
       "why": "The key acts as you on your servers."
      },
      {
       "dont": "Retrying the same call after invalid_args",
       "fix": "Fix the exact paths the error names, then call again",
       "why": "The error lists what to correct; the same call fails again."
      }
     ]
    }
   ]
  },
  {
   "id": "faq",
   "nav": "Questions",
   "h2": "Questions",
   "parts": [
    {
     "t": "faq",
     "items": [
      {
       "q": "Can the assistant run any command it wants?",
       "a": "No. It chooses from declared actions with typed inputs; the server program rejects unknown actions and inputs.",
       "evidence": {
        "url": "https://github.com/AndrewDryga/emisar/blob/31f4c55b3f7513621e09e3f4468ff29dfdc6e123/README.md",
        "quote": "The runner rejects unknown actions and arguments."
       }
      },
      {
       "q": "Who approves risky actions?",
       "a": "A person, in the emisar console. Approvals are never decided inside the assistant's chat.",
       "evidence": {
        "url": "https://github.com/AndrewDryga/emisar/blob/31f4c55b3f7513621e09e3f4468ff29dfdc6e123/.agent/kb/specs/mcp-api.md",
        "quote": "approvals are decided by an operator in the console, never by the model's session."
       }
      },
      {
       "q": "Does my server need an open port?",
       "a": "No. The program on the server only connects outward over an encrypted connection and listens for nothing.",
       "evidence": {
        "url": "https://github.com/AndrewDryga/emisar/blob/31f4c55b3f7513621e09e3f4468ff29dfdc6e123/README.md",
        "quote": "The runner opens an outbound TLS WebSocket and exposes no inbound listener."
       }
      },
      {
       "q": "Is my data used to train AI?",
       "a": "emisar says it does not use your data to train AI models.",
       "evidence": {
        "url": "https://emisar.dev/privacy",
        "quote": "We do not use your data to train AI models."
       }
      },
      {
       "q": "Is it free?",
       "a": "Free for 3 servers and 1 user with no card; Team is $20 per server a month.",
       "evidence": {
        "url": "https://emisar.dev/pricing",
        "quote": "Create a free account. No credit card required."
       }
      }
     ]
    }
   ]
  }
 ],
 "glance": [
  [
   "Registry name",
   "dev.emisar/emisar"
  ],
  [
   "Category",
   "Other"
  ],
  [
   "Version",
   "0.60.0"
  ],
  [
   "Remote endpoint",
   "https://emisar.dev/api/mcp/rpc"
  ],
  [
   "License",
   "Custom license"
  ],
  [
   "GitHub",
   "andrewdryga/emisar"
  ],
  [
   "In the registry since",
   "Nov 18, 2026"
  ],
  [
   "Works with",
   "Claude Code, OpenAI Codex, OpenCode, LangChain, n8n, LangGraph"
  ]
 ],
 "good": [
  "The program on each server runs allowed actions with its own system permissions and does not sandbox them, so an approved destructive action really happens.",
  "An optional shell pack gives full command access for emergencies; it is denied by default and should never be installed on production servers.",
  "The free plan covers 3 servers, 1 user and 7 days of audit history; Team costs $20 per server a month."
 ],
 "checked_line": "Read by AgentGid from 14 pages and files at one commit, checked Nov 10, 2026; stars update daily.",
 "logo": "https://agentgid.com/media/avatars/andrewdryga.png"
}