{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_q49mk9ue56dw","handle":"clops-mcp","url":"https://wellknown.network/agents/clops-mcp","links":{"self":"https://wellknown.network/agents/clops-mcp/record.json","html":"https://wellknown.network/agents/clops-mcp","markdown":"https://wellknown.network/agents/clops-mcp/record.md","api":"https://wellknown.network/api/v1/agents/clops-mcp","status":"https://wellknown.network/api/v1/agents/clops-mcp/status","claim":"https://wellknown.network/agents/clops-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/clops-mcp/claim.json","badge":"https://wellknown.network/agents/clops-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/clops-mcp/history","tools":"https://wellknown.network/api/v1/agents/clops-mcp/tools"},"ard":{"identifier":"urn:air::server:clops-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"clops-mcp","summary":"Long-running, multi-step Claude Code workflows, written in Python and run one focused step at a time.","description":"<!-- mcp-name: io.github.clops-mcp/clops-mcp -->\n\n# clops\n\nLong-running, multi-step Claude Code workflows, written in Python and run one\nfocused step at a time.\n\n## The problem\n\nYou write a skill for a workflow you run every week. It works. Then you add the\nedge cases, the exceptions, the \"don't forget to check X\" — and it stops\nworking. Claude cheats the steps it finds boring. It forgets details buried in\nthe middle. It overcorrects on whatever you emphasised last. The more detail you\nadd to make it reliable, the less reliable it gets, because the entire document\nis in context the entire time and Claude has to keep deciding, unprompted, which\npart applies right now.\n\nclops inverts that. Each step of the workflow is a separate **Op** with its own\nprompt, its own input, and its own subagent. The runtime decides which Op runs\nnext and renders the prompt for it. The agent doing the work sees one step's\nworth of instruction — not the workflow.\n\nThe whole goal of clops is to help Claude think about what it needs to think\nabout right now, and not a whole lot else.\n\nThe payoff compounds: a workflow you get right once stays right, and you can\ncompose it into bigger ones instead of rewriting it. Your workflows stop being\ndisposable.\n\n## What this is not\n\n**This is not an agent framework.** clops has no API key, no provider SDK, and\nno model calls of its own. It runs inside Claude Code, which supplies the model\nand the subagents. If what you want is durable execution, retries, and\nobservability for unattended production agents, you want LangGraph or Temporal,\nnot this.\n\n**This is not for unattended work.** clops targets interactive and\nsemi-interactive workflows — the ones where you are in the loop, or will be\nshortly. The value is *preparation*: the machine spends twenty minutes building\ncontext, and when your attention arrives the work is ready to go. If nobody is\ncoming back, clops buys you nothing.\n\n## Why not just write a skill?\n\n| Objection | Answer |\n|---|---|\n| \"Just…","publisher":{"name":"Wesley Harding","url":null},"homepage":"https://github.com/clops-mcp/clops-mcp/tree/main/docs","repository":"https://github.com/clops-mcp/clops-mcp/releases","version":"0.4.5","license":null,"protocols":["mcp"],"tags":["mcp","model-context-protocol","claude","claude-code","llm","agents","orchestration","pipeline","workflow"],"pricing":null,"endpoints":[{"url":"pypi:clops-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"dev.ci-cd","name":"CI/CD & Deploy","confidence":1,"provenance":"declared"},{"slug":"automation.orchestration","name":"Agent Orchestration","confidence":1,"provenance":"declared"},{"slug":"productivity.crm","name":"CRM & Sales","confidence":0.675,"provenance":"derived"}],"categories":["automation","dev","productivity"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"clops-mcp","registry":"pypi","observedAt":"2026-10-27T21:24:56.836Z","publishedAt":"2026-08-16T19:27:43.295447Z","latestVersion":"0.4.5"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"clops-mcp","url":"https://pypi.org/project/clops-mcp/","firstSeenAt":"2026-09-09T11:31:07.636Z","fetchedAt":"2026-10-27T21:24:35.347Z","normalizedAt":"2026-10-27T21:24:35.347Z"}]},"firstSeenAt":"2026-09-09T11:31:07.636Z","updatedAt":"2026-10-27T21:24:56.836Z"}