agent-harness GitHub
Menu

Overview

agent-harness Agent Harness Blueprint Blueprint plugin product brief The Harness Book Deny-floor known limitations Licensing scope and decision Replay tool product brief Agent Harness roadmap Harness Specs

Docs

Advisory agent-operations measurements Mapped authority-document drift probe Mapped authority drift: pinned local trial Benchmarks and measured evidence Corpus path boundary Advisory entropy measurements Learning evaluator evidence boundaries Outcome stream validation System comparisons and canary evidence Advisory logical-job accounting Replay lab Python check inventory Replay CLI input identity and timeout limits Replay input decoding failures Replay report output boundary Replay process policy-root aliases Replay supervisor marker scope Review evidence and acceptance oracles Offline reviewer-packet diagnostics System state Process-fixture startup is not containment evidence

Agentic

Architecture additives — perfect assistant → estate owners Perfect autonomous agentic assistant — estate synthesis

Evals

Harness extension notes Evaluation merge-boundary contract: v0 Merge-gate model Outcome ledger Review under flood Agentic development evaluation taxonomy: v0

Evals / Examples

Worked evidence packet: a private product PR

Extraction

Replay v0 extraction inventory

Freeze

Legacy deny-floor v1 freeze record

Maintenance

Managed Codex agent publication Sync input and Doctor session roots

Observability

Coding-agent loop observability v0 Offline source-record identity inspector

Observability / Adapters

Claude Code to agent-loop v0: mapping note

Review evidence

Review/oracle wave: source ledger

UX evaluation

Frozen observation verifier Evidence-bound issue seeds Advisory product UX evaluation Offline scenario binder Scenario authoring contract v0 Taskdeck-first adapter sketch Workspace handoff acceptance profile Workspace receipt lab

Harness extension notes

Status: proposed, 2026-09-23. Parent: #299. Spike: #302.

Extend the existing workbench

Review evidence and its v0 template remain canonical. The #301 flood profile is a review view, not another receipt authority.

Proposed paths: docs/evals/{TAXONOMY,MERGE_BOUNDARY,REVIEW_UNDER_FLOOD}.md (#300/ #304/#301); docs/observability/AGENT_LOOP_SPANS.md (#303) owns trace vocabulary. Use issue links until those files land.

Read-only adapter contracts, not implementations: PR/base/head/stack references map to identity; deterministic run/fixture references map to retained observations; expected check IDs map to observed/missing evidence; optional trace IDs map to advisory diagnostics. Unknowns and prior attempts survive projection. Packet commands remain data.

Before implementing, reconcile the pending #295 reader and #296 accounting contracts. No pending API is promised. Reader diagnostics retain execution_verified=false and merge_verdict=null. A private control-plane issue indexes evidence, never promotes authority. Only executed measurements enter BENCHMARKS.

Sequence and exclusions

Review #301, then this note; curate the real cohort separately. #295/#296/ #298 Verify-red work does not block docs. Do not change CI or waive merge requirements.

Defer the private skills/workflow packaging issue until source-of-truth docs and packet mapping stabilize. No packaging PR now.

No second harness, runner, collector, opentelemetry-sdk, runtime-schema migration, private product PR, product rewrite, UX scoring, or LLM merge gate.