{
  "listings": [
    {
      "slug": "agentos",
      "name": "AgentOS",
      "company": "Framers Lab, Inc.",
      "grade": "A",
      "fit": "caveat",
      "why": "SqlCheckpointStore (SQLite or Postgres) ships, and resume marks write and external nodes complete with their recorded output while failed nodes re-run. The docs don't cover resume across code changes or what a cancel keeps.",
      "source": "https://docs.agentos.sh/features/checkpointing",
      "confidence": "low",
      "url": "https://www.anchorterminal.com/tools/agentos"
    },
    {
      "slug": "agno",
      "name": "Agno",
      "company": "Agno Inc.",
      "grade": "BB",
      "fit": "caveat",
      "why": "Runs persist to SQLite or Postgres and continue by run_id in a new process. I found no rule for a tool call in flight at a crash, and resume across code changes isn't documented.",
      "source": "https://docs.agno.com/hitl/user-confirmation.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/agno"
    },
    {
      "slug": "boxlang-ai",
      "name": "BoxLang AI",
      "company": "Ortus Solutions, Corp.",
      "grade": "B",
      "fit": "caveat",
      "why": "A checkpointer memory (cache or jdbc) stores suspended state and cancelRun stops at the next beforeLLMCall or beforeToolCall. No documented rule for a tool in flight at a crash or resume across code changes.",
      "source": "https://ai.ortusbooks.com/main-components/agents/memory.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/boxlang-ai"
    },
    {
      "slug": "claude-agent-sdk",
      "name": "Claude Agent SDK",
      "company": "Anthropic, PBC",
      "grade": "BB",
      "fit": "caveat",
      "why": "Sessions persist and resume by id, and a SessionStore mirrors transcripts to your own store so a new container can hydrate. I found no rule for a tool call in flight when the process died; state on the container disk is lost on restart without a store.",
      "source": "https://code.claude.com/docs/en/agent-sdk/hosting.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/claude-agent-sdk"
    },
    {
      "slug": "crewai",
      "name": "CrewAI",
      "company": "CrewAI, Inc.",
      "grade": "C",
      "fit": "caveat",
      "why": "Checkpointing ships JsonProvider and SqliteProvider; a restore skips completed tasks and rehydrates memory. No Postgres or Redis store, no rule for an in-flight tool call, and lambdas and bound-method callbacks are not restored.",
      "source": "https://docs.crewai.com/v1.15.41/en/concepts/checkpointing.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/crewai"
    },
    {
      "slug": "cursor-agent-sdk",
      "name": "Cursor Agent SDK",
      "company": "Anysphere, Inc.",
      "grade": "B",
      "fit": "caveat",
      "why": "Sessions keep an append-only record of messages and tool calls that survives a restart, stop records turn.failed as cancelled, and tool effects mark writes run-once. The store is not named and resume across code changes isn't covered.",
      "source": "https://cursor.com/docs/agent-sdk/reference/sessions.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/cursor-agent-sdk"
    },
    {
      "slug": "docker-agent",
      "name": "Docker Agent",
      "company": "Docker, Inc.",
      "grade": "BB",
      "fit": "caveat",
      "why": "Every message, tool call and sub-agent run is stored in a SQLite session.db and a run resumes by session id. No rule for a tool call in flight at a crash, and resuming mid-run differs from continuing a conversation.",
      "source": "https://docker.github.io/docker-agent/features/sessions/",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/docker-agent"
    },
    {
      "slug": "google-adk",
      "name": "Agent Development Kit (ADK)",
      "company": "Google LLC",
      "grade": "BB",
      "fit": "caveat",
      "why": "ResumabilityConfig logs completed steps and reinstates tool results, re-running the interrupted tool so tools run at least once; resume after modifying the workflow is not supported, and resume isn't available from ADK Web or the CLI.",
      "source": "https://adk.dev/runtime/resume/index.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/google-adk"
    },
    {
      "slug": "guilds-run",
      "name": "guilds.run",
      "company": "Cigale AI",
      "grade": "C",
      "fit": "caveat",
      "why": "Runs live in PostgreSQL; on restart, unfinished committed tool calls get an interrupted result, pending group dispatch resumes, and routines wake agents on schedules. Resume across a code change isn't documented, and a run that hit its tool-call cap can't be resumed.",
      "source": "https://docs.guilds.run/architecture/orchestrator",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/guilds-run"
    },
    {
      "slug": "langgraph",
      "name": "LangGraph",
      "company": "LangChain, Inc.",
      "grade": "BB",
      "fit": "caveat",
      "why": "Postgres, SQLite and MongoDB checkpointers ship; checkpoints land at super-step boundaries and resume re-runs the node at least once, with idempotency-key guidance. Accounts: cancelling a run loses state since the last checkpoint (langgraph #5672, open) and Redis resume failed (#5074).",
      "source": "https://docs.langchain.com/oss/python/langgraph/checkpointers.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/langgraph"
    },
    {
      "slug": "llamaindex",
      "name": "LlamaIndex Framework",
      "company": "LlamaIndex, Inc.",
      "grade": "BB",
      "fit": "caveat",
      "why": "Context.to_dict() snapshots a run for any store, resume re-dispatches pending events and re-runs a step that was mid-execution (at-least-once), and a DBOS integration exists. You write the store; an account hit serialisation errors for Gemini function calls in Postgres (llama_index #19992).",
      "source": "https://developers.llamaindex.ai/python/llamaagents/workflows/durable_workflows/",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/llamaindex"
    },
    {
      "slug": "mastra",
      "name": "Mastra",
      "company": "Kepler Software, Inc.",
      "grade": "A",
      "fit": "caveat",
      "why": "Durable agents run the loop inside a workflow; with Inngest each tool call is a memoized step retried independently. Storage is libSQL, PostgreSQL or MongoDB. Agent run snapshots are deleted when the run finishes, and resume across code change isn't covered.",
      "source": "https://mastra.ai/docs/harness/durable-agents.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/mastra"
    },
    {
      "slug": "microsoft-agent-framework",
      "name": "Microsoft Agent Framework",
      "company": "Microsoft Corporation",
      "grade": "A",
      "fit": "caveat",
      "why": "Workflow checkpoints at superstep ends go to in-memory, file or Cosmos DB storage, and the Durable Task extension keeps agent sessions across crashes for days. Upgraded workers need replay-compatibility checks, and agent-id changes broke resume for one account (#4793).",
      "source": "https://learn.microsoft.com/en-us/agent-framework/workflows/checkpoints",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/microsoft-agent-framework"
    },
    {
      "slug": "openai-agents-sdk",
      "name": "OpenAI Agents SDK",
      "company": "OpenAI OpCo, LLC",
      "grade": "AA",
      "fit": "caveat",
      "why": "Sessions ship for SQLite, SQLAlchemy, Redis and Dapr, and a RunState serialises with to_string() to resume an interrupted or cancelled run. No rule for a tool call in flight at a crash was found.",
      "source": "https://openai.github.io/openai-agents-python/sessions/",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/openai-agents-sdk"
    },
    {
      "slug": "pydantic-ai",
      "name": "Pydantic AI",
      "company": "Pydantic Services Inc.",
      "grade": "AA",
      "fit": "caveat",
      "why": "Nine durable engines (Temporal, DBOS, Prefect, Restate, AWS Lambda and more) journal model and tool calls so a run survives a crash mid-step, and StepPersistence saves, resumes and forks runs. The overview doesn't cover resume across code changes or what a cancel keeps.",
      "source": "https://pydantic.dev/docs/ai/capabilities/durable_execution/overview/index.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/pydantic-ai"
    },
    {
      "slug": "strands-agents",
      "name": "Strands Agents",
      "company": "Amazon Web Services, Inc.",
      "grade": "A",
      "fit": "caveat",
      "why": "Session managers persist to local disk or S3 (SnapshotSessionManager recommended), with snapshot_latest used to resume after a restart. No Postgres or Redis store and no rule for a tool call in flight; the in-memory history gap of strands #33 is closed.",
      "source": "https://strandsagents.com/docs/user-guide/sdk/agents/session-management/index.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/strands-agents"
    },
    {
      "slug": "tradingagents",
      "name": "TradingAgents",
      "company": "Tauric Research",
      "grade": "B",
      "fit": "caveat",
      "why": "Opt-in --checkpoint saves LangGraph state after each node to a per-ticker SQLite database so a crashed run resumes from the last step. No rule for an in-flight tool call; its tools only read data, so repeats have no side effect.",
      "source": "https://raw.githubusercontent.com/TauricResearch/TradingAgents/main/README.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/tradingagents"
    },
    {
      "slug": "vercel-ai-sdk",
      "name": "AI SDK",
      "company": "Vercel Inc.",
      "grade": "AA",
      "fit": "caveat",
      "why": "WorkflowAgent from @ai-sdk/workflow runs each tool call as a durable workflow step with retries and survives restarts; ToolLoopAgent itself is in memory. Durability depends on Vercel Workflow, and WorkflowAgent applies no default step limit.",
      "source": "https://ai-sdk.dev/docs/agents/workflow-agent",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/vercel-ai-sdk"
    },
    {
      "slug": "nitrostack",
      "name": "NitroStack",
      "company": "Nitrostack Inc.",
      "grade": "C",
      "fit": "no",
      "why": "MCP Tasks cover long-running tool calls with progress, cancellation and a TTL, but there is no agent run state or checkpoint store; an account saw tasks still working past their TTL deleted (nitrostack #329, open).",
      "source": "https://docs.nitrostack.ai/docs/sdk/tasks.md",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/nitrostack"
    },
    {
      "slug": "smolagents",
      "name": "smolagents",
      "company": "Hugging Face, Inc.",
      "grade": "B",
      "fit": "no",
      "why": "No checkpoint store or resume page in the docs index; an open account proposes state serialisation because none exists (smolagents #364).",
      "source": "https://huggingface.co/docs/smolagents/llms.txt",
      "confidence": "medium",
      "url": "https://www.anchorterminal.com/tools/smolagents"
    }
  ],
  "marks": "yes: the docs show it does the job as the success test says. caveat: it does the job with a limit an agent should know. no: it doesn't do this job. unchecked: we couldn't read it.",
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "attributionRequired": true,
    "credit": "CC BY 4.0, credit Anchor Terminal and link the page or API address you took it from (https://www.anchorterminal.com/terms/#data)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-11-16",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.8",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run",
    "terms": "https://www.anchorterminal.com/terms/#data"
  },
  "useCase": {
    "accounts": 9,
    "after": [
      "human-in-the-loop.durable-wait",
      "agent-observability.debug-failed-run"
    ],
    "area": "frameworks",
    "asks": [
      "run this report for every customer each night",
      "resume the agent where it stopped after a crash",
      "keep the conversation after the server restarts",
      "cancel a run and carry on the thread"
    ],
    "before": [
      "agent-harnesses.scheduled-job"
    ],
    "blockers": [
      "only an in-memory or file checkpoint store ships",
      "history with provider-specific function-call objects fails to serialise to the store",
      "resume fails when agent ids or code change between runs",
      "a cancelled run loses everything after the last checkpoint"
    ],
    "capabilities": [
      "agent.framework",
      "agent.durable"
    ],
    "category": "frameworks",
    "categoryName": "Agent frameworks \u0026 SDKs",
    "checked": "2026-11-15",
    "companies": {
      "fit": 0,
      "fitWithLimit": 18,
      "tagged": 25
    },
    "competition": "none",
    "fits": null,
    "fitsWithLimit": [
      "agentos",
      "agno",
      "boxlang-ai",
      "claude-agent-sdk",
      "crewai",
      "cursor-agent-sdk",
      "docker-agent",
      "google-adk",
      "guilds-run",
      "langgraph",
      "llamaindex",
      "mastra",
      "microsoft-agent-framework",
      "openai-agents-sdk",
      "pydantic-ai",
      "strands-agents",
      "tradingagents",
      "vercel-ai-sdk"
    ],
    "id": "frameworks.durable-long-run",
    "listings": {
      "caveat": 18,
      "no": 2,
      "notRead": 0,
      "tagged": 25,
      "unchecked": 0,
      "yes": 0
    },
    "name": "Run a long or scheduled agent job that survives restarts",
    "status": "draft",
    "success": "State is written to a named durable store after each step, a run killed mid-way resumes from the last completed step with a documented rule for the tool call that was in flight, cancelling keeps what was already streamed, and the docs show resume across a code change or new agent ids. Measured: kill the process during step 3 of 5 and count repeated tool calls on resume.",
    "url": "https://www.anchorterminal.com/api/v1/use-cases/frameworks.durable-long-run.json",
    "weight": 11,
    "weightBasis": "7 of 61 accounts have this as their main job (share 11.5, rounded by largest remainder to sum 100)"
  }
}
