Skip to content

Releases: aiming-lab/AutoHarness

v0.1.0 — The Harness Engineering Framework for AI Agents

02 Apr 00:27
@Jiaaqiliu Jiaaqiliu
v0.1.0
aab917c

Choose a tag to compare

View all tags
v0.1.0 — The Harness Engineering Framework for AI Agents Latest
Latest

AutoHarness v0.1.0

Three-tier governance pipeline (Core / Standard / Enhanced), 6-step tool governance, risk pattern matching, YAML constitution, trace-based diagnostics, multi-agent profiles, context management, session persistence with cost tracking.

958 tests passing.

Highlights

  • Pipeline Modes: Core (6-step), Standard (8-step), Enhanced (14-step, default)
  • Risk Patterns: Dangerous commands, secret detection, path traversal, code injection
  • Multi-agent: Role-based profiles (coder / reviewer / planner / executor)
  • Context: Token budget, microcompact, LLM summarization, image stripping
  • CLI: autoharness init, mode, check, validate, audit, install, export
  • 2 lines to integrate: AutoHarness.wrap(OpenAI())

Install

git clone https://github.com/aiming-lab/AutoHarness.git
cd AutoHarness && pip install -e .
Assets 2
Loading