HarnessMatch
Menu

Coding agent

Qwen Code

Open coding agent with persistent memory, delegation, and opt-in isolation.

Status
Active
License
Apache-2.0
Evidence
Documented + code-verifiable, 17 sources
Product record checked
2026-07-27
Compare harnessesInspect sources

At a glance

An Apache-2.0 terminal coding agent with persistent auto-memory, headless execution, hosted and local providers, MCP, specialized subagents, built-in desktop Computer Use, IDE integrations, checkpoints, and opt-in macOS or container isolation.

Good choice if

  • Terminal and CI workflows that can explicitly enable a first-party sandbox path
  • Teams switching among Qwen, OpenAI-compatible, Anthropic, Gemini, and local endpoints
  • Delegated coding tasks that benefit from subagents, MCP, skills, and IDE integrations

Check before choosing

  • The sandbox is off by default; `--yolo` does not enable it, and the default macOS profile allows network access while mainly restricting writes
  • Checkpointing is also opt-in and restores tracked project state and conversation, not shell-created external side effects
  • Auto-memory is enabled by default and can persist user or project knowledge, so teams should review what is retained and synchronized
See 1 more considerations
  • The qwen serve daemon is an alpha local API: loopback starts without authentication, non-loopback binding fails closed without a bearer token, and the docs do not yet claim production-grade multi-client or network-failure guarantees

Why it qualifies as a coding harness

This confirms category fit, not product quality. Every required criterion links back to first-party evidence.

Qualifies4 of 4 required criteria evidenced

  • Adaptive agent loop

    Documented

    The system repeatedly observes results and chooses the next action instead of following a fixed one-pass graph.

    EvidenceQwen Code overview

  • Repository tool execution

    Documented

    The system can use tools to inspect and change a repository or its execution environment.

    EvidenceQwen Code 0.21.0 source snapshot

  • Task-aware context management

    Documented

    The runtime assembles, updates, compacts, retrieves, or persists task-relevant context while work proceeds.

    EvidencePersistent memory

  • Model-independent runtime control

    Documented

    Permissions, budgets, interruption, policy, or stop controls operate outside the model's own text generation.

    EvidenceApproval and settings reference

Membership establishes category fit only. It does not score quality, safety, autonomy, model capability, or benchmark performance. · Read the membership rule.

How it works under the hood

Seven mechanisms mapped from first-party records. These labels describe what the harness provides, not how intelligent its model is.

7/7layers documented
  • Execution & isolationSandbox availableDocumented mechanism, not a performance score.
  • Tooling & integrationsExtensible + browserDocumented mechanism, not a performance score.
  • Context & statePersistent stateDocumented mechanism, not a performance score.
  • Lifecycle & recoveryCheckpoint/rewindDocumented mechanism, not a performance score.
  • ObservabilityStructured tracesDocumented mechanism, not a performance score.
  • VerificationTool-assistedDocumented mechanism, not a performance score.
  • Governance & permissionsPolicy controlsDocumented mechanism, not a performance score.

Public code audit

5/5public artifacts present
Security policy
Present at inspected commit
CI workflow
Present at inspected commit
Automated tests
Present at inspected commit
Evaluation assets
Present at inspected commit
Contributor documentation
Present at inspected commit

The v0.21.0 tree contains 2,062 test-like files, 39 workflows, security and contributor documentation, and one codegraph eval fixture. The fixture is project-owned development evidence rather than an immutable comparative result; provider services and hosted model behavior remain outside the source audit.

Inspect commit 5610eb405212f807a482214ddd28a259da7855d3, checked 2026-07-27

Measured configurations

No benchmark run passes the full metadata admission policy for this harness yet. Missing data is not scored as zero.

Benchmark policy and all runs

Capability support

Documented first-class product support, checked against the sources below.

External tools (MCP)
DocumentedProduct-supported MCP integrationSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Local models
DocumentedLocal or self-hosted model pathSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Agent parallelism
DocumentedDelegated or parallel agent workflowSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Runs without an open UI
DocumentedNon-interactive or automation surfaceSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Browser control
DocumentedBuilt-in or product-supported browser controlSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Isolated execution
OptionalOptional OS or container sandboxSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.
Undo file changes
DocumentedProduct-supported file or session rollbackSource · checked 2026-07-27The source establishes the mechanism, not its quality or availability in every mode.

Primary evidence

Each capability claim is tied to a first-party record and a verification date.

Product record checked 2026-07-27
Qwen Code 0.21.0 source snapshotImmutable Apache-2.0 source snapshot for the audited stable release, including CLI, desktop, tests, workflows, and one project-owned codegraph eval fixtureOfficial repositoryQwen Code 0.21.0 releaseStable release identity and dated version boundary used for the auditOfficial announcementSandboxOpt-in macOS Seatbelt, Docker and Podman isolation; provider selection; filesystem and network profiles; and the network-open macOS defaultOfficial docsApproval and settings referenceDefault approval for edits and shell, plan, auto-edit, classifier-based auto mode, and unrestricted yolo modeOfficial docsHeadless safety and budgetsStructured automation, explicit run budgets, and the warning that yolo neither enables nor substitutes for a sandboxOfficial docsPersistent memoryQWEN.md instructions, default-on auto-memory, user and project scopes, review commands, cleanup, and optional Git synchronizationOfficial docsCheckpointingDisabled-by-default shadow-Git snapshots, local conversation capture, restore behavior, and configurationOfficial docsCommands and recoverySession resume and branching, conversation rewind, checkpoint restore, exports, budgets, and background-agent commandsOfficial docs
View 9 additional sources
Product and interfaces4 sources
Built-in Computer Use releaseZero-configuration desktop Computer Use as a native capability rather than an inferred browser feature from WebFetch or MCPOfficial announcementModel providersOpenAI-compatible, Anthropic, Gemini, Qwen, Vertex AI, and local self-hosted model configurationOfficial docsQwen Code overviewOpen-source agent scope, terminal workflow, supported integrations, installation entry points, and official documentation mapOfficial docsJetBrains integrationJetBrains IDE connection through the Agent Client Protocol, setup, session interaction, and editor context exchangeOfficial docs
Automation and extensions3 sources
MCP server integrationModel Context Protocol server configuration, transport and tool discovery, lifecycle management, and extension pointsOfficial docsQwen Code ExtensionsInstallable extension packaging for agents, skills, MCP servers, context files, workspace scopes, enablement, and removalOfficial docsqwen serve daemonAlpha local HTTP and event-stream API, Web Shell, sessions, persistence, loopback and remote authentication, TLS, and maturity limitsOfficial docs
Enterprise and operations1 sources
OpenTelemetry observabilityDefault-off local or OTLP logs, metrics, spans, tool and subagent events, sensitive-data controls, and outbound trace propagation boundariesOfficial docs
Releases and public code audit1 sources
Nested subagent releaseDated release notes for nested subagent spawning, default depth controls, tree visibility, parameter-level permissions, and Web Shell sessionsOfficial announcement