Coding agent

Spectral Agent

Local coding harness with goal loops, managed context, subagents, and browser relay access.

Status
Active
License
MIT (published npm package)
Evidence
Documented, 7 sources
Product record checked
2026-08-08

At a glance

A local coding harness from Aexol that reads and writes repository files, runs shell and Git commands, iterates toward explicit goals, compacts and recalls session context, delegates work to isolated-context subagents, and connects either through a terminal UI or a browser relay. Execution is host-first: the reviewed documentation does not establish an operating-system sandbox or a general tool-permission policy.

Good choice if

  • Repository tasks that benefit from explicit goal loops with bounded iterations and clean user interruption
  • Long terminal or browser-relay sessions using automatic compaction, recall, and locally persisted session history
  • Delegating focused work to project or user-defined subagents with separate context, filtered tools, and optional model overrides

Check before choosing

  • The npm package declares an MIT license and a GitLab repository, but that repository currently requires sign-in, so HarnessMatch could not pin and inspect the implementation
  • File, shell, Git, extension, and local MCP tools execute on the host; the reviewed documentation establishes no operating-system sandbox or general allow, ask, and deny policy
  • Project-local subagent definitions prompt for confirmation in interactive local use, but that confirmation is bypassed in headless or relay mode
See 3 more considerations
  • The relay routes browser traffic and stores identity metadata; in relay mode the inference proxy holds provider keys, while local TUI mode stores keys in the local Spectral configuration
  • Loop completion is decided inside the agent from progress toward the goal, not by an independent correctness gate; iteration, time, and spend limits remain necessary
  • Compaction, recall, and session reconnection preserve conversational work but do not establish file rollback or reversal of shell, Git, network, or external-service side effects

Capability support

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

External tools (MCP)
DocumentedProduct-supported MCP integrationSource · checked 2026-08-08The source establishes the mechanism, not its quality or availability in every mode.
Reusable skills
Not documentedNo first-class support established by the current recordAbsence of current documentation is not proof that the capability is impossible.
Local models
Not documentedNo first-class support established by the current recordAbsence of current documentation is not proof that the capability is impossible.
Agent parallelism
DocumentedDelegated or parallel agent workflowSource · checked 2026-08-08The source establishes the mechanism, not its quality or availability in every mode.
Runs without an open UI
Not documentedNo first-class support established by the current recordAbsence of current documentation is not proof that the capability is impossible.
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
Not documentedNo first-class support established by the current recordAbsence of current documentation is not proof that the capability is impossible.
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 @aexol/spectral 0.9.152, authenticate, choose local TUI or browser-relay mode, define a project directory, review subagent and MCP configuration, and add external isolation before running untrusted or unattended work.

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 & isolationHost processDocumented mechanism, not a performance score.
  • Tooling & integrationsExtensible toolsDocumented mechanism, not a performance score.
  • Context & stateManaged contextDocumented 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 & permissionsHost accessDocumented 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

No official public repository was located for a code-level audit.

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
  • npm last-month downloads13.14KPackage: @aexol/spectralOpen 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.

7 first-party sourcesProduct record checked

Product and interfaces

1 source

Agents, state and recovery

4 sources
View 3 more sources

Automation and extensions

1 source

Releases and public code audit

1 source

Ecosystem context

OpenRouter Aexol Studio app profile Discovery only: the app identifies as Aexol Studio, so its usage is not attributed to Spectral Agent. Membership and capabilities are established from Aexol documentation and immutable package metadata. Observed 2026-08-08.