Coding agent

CodeWhale

Model-portable Rust coding agent with governed execution and resumable fleets.

Status: Active
License: MIT
Evidence: Documented + code-verifiable, 11 sources
Product record checked: 2026-08-02

At a glance

An MIT-licensed terminal coding harness with an adaptive tool loop, interactive and headless modes, persistent memory, turn restore, MCP and skills, local-model routes, optional OS or external sandboxes, and ledger-backed multi-worker fleets.

Good choice if

Terminal or CI workflows that need the same tools across hosted, local, and OpenAI-compatible model routes
Teams that want explicit planning, approval, repository-law, and optional OS-sandbox controls
Longer tasks that benefit from persistent memory, turn restore, and ledger-backed multi-worker resume

Check before choosing

Local execution is host-first: macOS Seatbelt depends on a successful probe, Linux bubblewrap is opt-in, Windows has no advertised local OS sandbox, and unavailable wrappers fall back unless external policy blocks the run
The strongest approval and lifecycle-hook controls are configurable; Full Access weakens prompts, and TUI hooks do not fire for codewhale exec or other CLI subcommands
The loopback web client is a local control surface rather than remote managed execution, and its one-time token does not turn it into a multi-user service

See 2 more considerations

Fleet workers reuse headless CodeWhale and an append-only ledger; this establishes coordination and resume, not independent quality or task-success evidence
Repository eval and acceptance assets are project-owned development evidence, so no product benchmark score is imported

Quick facts

Interfaces: Terminal, Web, Automation
Approval style: Policy rules
Model portability: Provider + local
Isolation: OS sandbox, Managed sandbox, Git worktree
Agent organization: Multi-agent runtime
Local model path: DocumentedThe source establishes the mechanism, not its quality or availability in every mode.

View full technical record

Catalog layer: Coding harness
Harness membership: Qualifies: 4 required criteria evidenced
Product role: Coding agent
Agent organization: Multi-agent runtime
Runtime posture: Host-first
Isolation options: OS sandbox, Managed sandbox, Git worktree
State model: Persistent memory
Context lifecycle: Persistent memory
Permission posture: Policy rules
Verification: Tool-assisted checks
Observability: Structured traces
Recovery: File checkpoint
Interfaces: Terminal, Web, Automation
Provider posture: Multi-provider
Model portability: Provider + local
Subscription access: Available
Enterprise access: Not documented
Local model path: DocumentedThe source establishes the mechanism, not its quality or availability in every mode.
Claim boundary: Harness capability only

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 macOS Seatbelt, opt-in Linux bubblewrap, or configured external OpenSandbox executionSource · checked 2026-08-02Unavailable local wrappers can fall back to host execution; Windows has no advertised local OS sandbox.
Undo file changes: DocumentedTurn-level restore in the interactive TUISource · checked 2026-08-02Turn restore does not establish rollback of external side effects.

Getting started

Install CodeWhale 0.9.3 from npm, Cargo, or a platform archive; select a provider and model; then choose permission, sandbox, memory, hook, and fleet settings for the workflow.

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

Adaptive agent loop
Documented
The system repeatedly observes results and chooses the next action instead of following a fixed one-pass graph.
EvidenceCodeWhale 0.9.3 source snapshot
Repository tool execution
Documented
The system can use tools to inspect and change a repository or its execution environment.
EvidenceCodeWhale 0.9.3 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.
EvidenceAuthorization order

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 & 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.

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

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 pinned tree contains the Rust harness implementation, 24 workflows, 111 test-like files, security and contribution guidance, and project-owned eval and acceptance assets. Those assets establish development practice rather than an independent, immutable model × harness × environment × budget × attempts result, so no product score is imported.

Inspect commit 4f2c97b0d75c039a9b6069ebcf210cc499583376, 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 releasev0.9.3Released 2026-07-31; 61 stable releases in 90 daysOpen release
npm last-month downloads17.75KPackage: codewhaleOpen artifact
Release asset downloads156.64K31 stable releases; 162 matched assetsOpen artifact
GitHub stars40.35KFull-source repository; 3.49K forksOpen artifact