agent-notes-toolkit
Run the following command in DeepSeek Harness:
dsh plugin install liangminhua/agent-notes-toolkit
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install liangminhua/agent-notes-toolkit in DeepSeek Harness to install this plugin, source at https://github.com/liangminhua/agent-notes-toolkit
About this plugin
In multi-agent workflows, decisions tend to live only in chat logs with no structured trace and no repeatable quality check, so CI cannot enforce note standards on agent output. Agent Notes Toolkit packages the "Agent Notes" mechanism as a portable tool: a decision-record tree as the skeleton, six verification gates (lifecycle closure, classification, format, archive, links, paragraph wrap), and a shared engine that powers the local CLI, CI pipelines, and in-session tool calls with the same gate code and the same exit-code contract.
Beyond verification, the toolkit ships a scaffolding CLI that drops rule files into your project in one step, ready-to-use GitHub Actions and GitLab CI workflow definitions, and a self-contained dsh bundle that registers a dedicated AN preset, a notes-verify model tool, and companion skills without requiring a global install.
It is aimed at small-to-medium teams that need auditable, version-controlled decision records across agent sessions and want to enforce note quality as a first-class concern in their CI pipeline, especially engineers already working with dsh or looking to bring agent workflows under version control.
Use Cases
- Run six structure and format gates on agent-generated notes in CI pipelines
- Scaffold Agent Notes rule files and CI workflows into a project with one CLI command
- Inject a verification tool into dsh sessions to check the decision-record tree in real time
Best For
- Small teams needing auditable decision records across multi-agent collaboration
- Engineers who want agent workflows under version control and CI enforcement
- dsh developers seeking unified verification logic across sessions and pipelines
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.