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

Architecture additives — perfect assistant → estate owners

Status: proposed documentation, 2026-10-01. Parent: #299. Companion: PERFECT_ASSISTANT.

Thin map only. Prefer extending existing docs/issues; do not fork tracks.

Pattern (external) Estate additive Home Issue Status
Tool-use loop + verify + budget Document loop contract; keep delegate-runtime L1 docs/agentic/PERFECT_ASSISTANT.md #299 This PR
Deterministic Oracle/Nop solvability Cite for flood sample design docs/evals/REVIEW_UNDER_FLOOD.md #301 Schema landed; sample open
FAIL_TO_PASS ∩ PASS_TO_PASS Eval bakeoffs / Bug Hunt sister docs/evals/TAXONOMY.md #325 sister Cross-link
Merge boundary lock Reuse the existing proposal docs/evals/MERGE_BOUNDARY.md #304 Landed proposal; gate adoption separate
Metadata-first OTel spans Adapter + native trial docs/observability/* #320 Docs landed; native NOT RUN
Review bots advisory Severity bar + packet docs/evals/REVIEW_UNDER_FLOOD.md #301 Landed schema
Session vs durable memory Persist ledger/decisions only delegate-runtime journal · harness ledger runtime journal issue / harness ledger PRs Cite
MCP isError + timeouts Reliability checklist (docs) this folder · later skill #299 / configuration-repository skill issue later Docs now; skill park
Human interrupt / approve Draft-first · decision@2 · estop the console · the merge gate console round 4 · control-plane issue Cite delegate-runtime pack
Coverage without gate gate_eligible: false control-plane evidence coverage-evidence issue (closed) Boundary cite
UX observe→judge Separate epic docs/ux-evaluation/ #281 Cross-link only
Design-grade playbooks Separate control-plane epic control-plane playbooks control-plane playbooks issue Cross-link only

Sequencing

  1. Land this docs pair (draft → human review → merge when ready).
  2. Run #320 content-off native trial on workstation.
  3. Finish #301 READY sample via #307 curation.
  4. Close #300 only after the ORACLES/PATTERNS agreement comment.
  5. Defer configuration-repository skill packaging until SoT docs stable (HARNESS_EXTENSION).

Non-goals

No CI changes · no opentelemetry-sdk · no second harness · no LLM merge gate · no #281 product work in this PR.