HarnessMatch

Search 59 product records by name, workflow, interface, or documented capability.

Quick access
Use ↑ ↓ to move, Enter to open, Esc to close.

Quick access links shown.

Menu

Extensible harness

OpenHarness

Python coding harness with skills, memory, governed tools, and agent teams.

Status
Active
License
MIT
Evidence
Documented + code-verifiable, 10 sources
Product record checked
2026-08-02
Compare harnessesInspect sources

At a glance

An MIT-licensed Python harness that owns a streaming tool loop, repository tools, compaction and persistent memory, permissions and hooks, optional OS or Docker sandboxes, provider and subscription routing, headless output, and multi-agent coordination.

Good choice if

  • Python users who want a Claude-style coding loop with portable skills, plugins, hooks, and MCP
  • Multi-provider workflows spanning API keys, local OpenAI-compatible endpoints, Copilot OAuth, and Claude or Codex subscription credentials
  • Interactive, scripted, scheduled, or delegated tasks that need compaction, persistent memory, and resumable sessions

Check before choosing

  • Sandboxing is disabled and fail-open by default; the OS wrapper requires Anthropic's separate sandbox-runtime package, and Docker isolation requires explicit configuration and a usable daemon
  • The permissive full-auto mode and dangerous skip flag bypass normal confirmations, while path and command rules remain policy controls rather than isolation
  • Subscription bridges reuse local Claude or Codex credentials as model routes; they do not transfer the capabilities, security posture, or UI of those external harnesses
See 2 more considerations
  • The bundled ohmo personal agent adds messaging gateways and autonomous repository work, but those surfaces should not be assumed for every OpenHarness embedding
  • The public repository has broad automated tests but no admitted independent product benchmark or repository security policy at the inspected commit

Capability support

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

External tools (MCP)
DocumentedProduct-supported MCP integrationSource · checked 2026-08-02The source establishes the mechanism, not its quality or availability in every mode.
Reusable skills
DocumentedProduct-supported reusable skill packagesSource · checked 2026-08-02Support does not establish portability, package quality, safety, or adoption.
Local models
DocumentedLocal or self-hosted model pathSource · checked 2026-08-02The source establishes the mechanism, not its quality or availability in every mode.
Agent parallelism
DocumentedDelegated or parallel agent workflowSource · checked 2026-08-02The source establishes the mechanism, not its quality or availability in every mode.
Runs without an open UI
DocumentedNon-interactive or automation surfaceSource · checked 2026-08-02The source establishes the mechanism, not its quality or availability in every mode.
Browser control
Not documentedNo first-class support established by the current recordAbsence of current documentation is not proof that the capability is impossible.
Isolated execution
OptionalOptional sandbox-runtime OS wrapper or per-session Docker containerSource · checked 2026-08-02Sandboxing and fail-closed behavior are disabled by default and depend on separately available runtimes.
Undo file changes
Not documentedNo first-class support established by the current recordAbsence of current documentation is not proof that the capability is impossible.

Getting started

Install openharness-ai 0.1.9, run oh setup to select an API, local, Copilot, Claude-subscription, or Codex-subscription profile, then configure permissions, optional sandboxing, memory, skills, MCP, and agent-team limits.

Open official documentation

Classification and operating model

Category fit and technical mechanisms are evidence records, not product-quality scores.

Category fit
Qualifies, 4/4 criteria
Operating model
7/7 layers documented
Inspect category criteria and operating mechanismsFirst-party records

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

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 toolsDocumented mechanism, not a performance score.
  • Context & statePersistent stateDocumented mechanism, not a performance score.
  • Lifecycle & recoverySession resumeDocumented mechanism, not a performance score.
  • ObservabilityLogs/transcriptsDocumented mechanism, not a performance score.
  • VerificationTool-assistedDocumented mechanism, not a performance score.
  • Governance & permissionsPolicy controlsDocumented mechanism, not a performance score.

Measured and public context

Configuration-specific measurements and source-native activity stay separate from general product capability.

Inspect code audit, measured configurations, and ecosystem signalsContext, not a product score

Public code audit

3/5public artifacts present
Security policy
Not found
CI workflow
Present at inspected commit
Automated tests
Present at inspected commit
Evaluation assets
Not found
Contributor documentation
Present at inspected commit

The pinned tree contains the Python harness and ohmo implementations, 117 test-like files, four workflows, release notes, and contributor guidance. No repository security policy or admissible independent product evaluation suite was found at the inspected commit.

Inspect commit 9b2efd795c6aa09f88b0c257d269a9e518da6ae7, checked 2026-08-02

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
Context, not quality

Public ecosystem signals

Source-native observations for exact mapped artifacts and reviewed stable release trains. Different units and populations stay separate, and missing coverage is never treated as zero.

View this harness in Usage
  • Latest stable releasev0.1.9Released 2026-05-07; 2 stable releases in 90 daysOpen release
  • Release asset downloads692 stable releases; 2 matched assetsOpen artifact
  • GitHub stars15.17KFull-source repository; 2.46K forksOpen artifact

Routing, package retrievals, release downloads, editor installs, and repository interest observe different populations. They are never added together and never affect capability evidence, classification, or measured results.

Interpretation rulesSignals checked

First-party evidence

Each capability claim links to the first-party record that supports it.

10 first-party sourcesProduct record checked

Product and interfaces

2 sources
View 1 more source

Execution and control

5 sources
View 4 more sources

Agents, state and recovery

1 source

Automation and extensions

1 source

Releases and public code audit

1 source