---

Table of Contents

• About The Project
• Current Status
• Features
• Architecture
• Getting Started
  • Prerequisites
  • Source Installation
• Usage
• Configuration
• Commands
• Session Events
• Documentation
• Roadmap
• Known Limitations
• Contributing
• License
• Acknowledgments

About The Project

Long AI-coding sessions get expensive — and drift. dsh-price-less is an unofficial context steward for DeepSeek Harness: it remembers what you are working on, tidies the context sent to the model — keeps what matters, drops what doesn't, and never touches anything it isn't sure about — so every token you send earns its place.

• Explicit tasks: /task tells the plugin what you are working on; opening and closing tasks is always your call
• Project brief (project frame): /init asks a few questions and writes a project brief, so the model always knows what this project is
• Intent detection: decides whether each message continues the current task or starts a new one — keyword rules first, a light auxiliary model only when unsure
• Prompt optimizer: /optimize-prompt tidies the current prompt on demand; the coming star button shares the same entry
• Log-only & replayable: every plugin action leaves a replayable fact log; on any failure it changes nothing — your content is never lost

Current Status

Area	Status
R1 platform foundation (events, storage, logs, LLM calls)	Completed
R2 detection & optimizer core (P8–P13)	Completed
P14a/P14b prompt optimizer + star button	Not yet wired
R3 shear (slim tool results) / R4 compaction (history compression)	Planned
npm/tarball distribution	Not published yet

Important: automatic intent detection is off by default. Enable it explicitly with discriminator.auto = true. The /optimize-prompt command is registered, but the full prompt-optimizer flow arrives with P14b.

Features

• Task boundaries: /task, /task close, /task status — split continuous work into clearly bounded tasks
• Project brief: /init proposal → your confirmation → persisted as project brief v1 (project_frame)
• Intent detection: your explicit instruction first, then zero-cost keyword rules, then past verdicts, and only then a light model — anything unresolved stays untouched (fail-lazy)
• Prompt optimizer: /optimize-prompt entry point, shared with the future star button
• Replayable facts: all context-economy/* events are log-only and ignorable
• Graceful degradation: if the harness lacks the ignorable channel, facts fall back to KV mirroring — still recorded, just stored elsewhere

Architecture

The diagram shows the current implemented architecture with color-coded layers and clearly marks planned modules as dashed. How to read it in one line: you and the harness sit on top; domains/ decides what to do, core/ does the computing, platform/ does the connecting — the bottom band is what gets recorded.

src/
├─ platform/    # adapter layer — the plugin's only way to touch the harness
├─ core/        # pure logic — computation only, zero harness imports
├─ domains/     # orchestration — commands, intent detection, wiring
├─ config.ts    # config schema and defaults
├─ settings.ts  # settings UI wiring
└─ index.ts     # plugin entry
client/         # settings card and star-button UI
docs/           # design canon and work orders

Getting Started

Prerequisites

• Node.js with ESM support (Node 20+ recommended)
• A local DeepSeek Harness source checkout for building
• dsh CLI or dev_inject_plugin for loading the plugin

Source Installation

This is currently a source-build / developer installation. There is no published npm package or tarball distribution yet.

# 1. Install dependencies
npm install --legacy-peer-deps --ignore-scripts --no-audit --no-fund --no-package-lock

# 2. Build from source (requires a local DSH checkout)
DSH_CHECKOUT=/path/to/deepseek-harness npm run build

# 3. Inject into a DSH instance
dev_inject_plugin /path/to/dsh-price-less

Planned distribution forms (not available yet):

• npm: dsh plugin add dsh-price-less
• tarball: dsh plugin add ./dsh-price-less-0.0.1.tgz

Usage

/init I am building a context management plugin
/init confirm

/task Design the command face
/task
/task close

Configuration

Configuration is currently grouped under discriminator.

Key	Type	Schema default	Description
discriminator.auto	boolean	false	Master switch for automatic intent detection
discriminator.provider	string	unset	Auxiliary LLM provider; when set together with model, overrides the built-in preset
discriminator.model	string	unset	Auxiliary LLM model; when set together with provider, overrides the built-in preset

The schema declares no default for provider/model. Left empty, auxiliary calls (intent detection, /init) fall back to the built-in preset deepseek-official / deepseek-v4-flash-vision-exp; setting both values overrides it. Enabling discriminator.auto or using /init may invoke the auxiliary LLM. This can incur API costs and may send relevant prompt content to the configured provider.

Commands

Command	Purpose
/task <description>	Explicitly open a new task
/task close	Close the current task
/task	Show current task status
/init <goal>	Start a project brief proposal
/init confirm	Confirm and persist the project brief v1 (project_frame)
/init cancel	Cancel the current proposal
/init	Show the current project brief
/optimize-prompt	Prompt optimizer entry; full flow pending P14b

Session Events

All currently emitted custom events are log-only and carry ignorable: true.

Event	Meaning
context-economy/task-boundary	Task boundary fact (which task opened / closed)
context-economy/judge-recorded	One complete intent-detection record
context-economy/judge-error	One failed detection record
context-economy/judge-verdict	The verdict: new task or not

Documentation

The design canon is maintained under docs/:

• 00 · System overview
• 01 · Architecture
• 02 · Discriminator
• 03 · Shear
• 04 · Compactor
• 05 · Constitution
• 06 · Cache
• 07 · Metrics
• 09 · State
• 10 · Wiring
• 11 · Structure
• 12 · Platform capabilities
• 13 · Harness plugin spec
• Implementation master plan

Roadmap

Phase	Scope	Status
R1	Platform foundation: events, storage, skills, LLM, history, tools	Completed
R2	Detection: task boundaries, project brief, intent detection & prompt optimizer	P8–P13 done; P14 pending
R3	Shear: slim tool results	Planned
R4	Compaction: history compression	Planned

See docs/implement/00-master.md for detailed work orders.

Known Limitations

• Not published: the package is still marked private and there is no official npm/tarball release.
• Intent detection is opt-in: discriminator.auto defaults to false to avoid unexpected auxiliary LLM costs.
• Prompt optimizer is incomplete: /optimize-prompt is only an entry point; the full star-button flow depends on P14b.
• R3/R4 not implemented: shear (slimming tool results) and compaction (history compression) are designed but not yet available.
• Local checkout required: building currently requires a local DeepSeek Harness source checkout and dev_inject_plugin.
• Ignorable-channel dependency: if the host harness does not include the local ignorable channel, plugin facts are mirrored to KV instead of being written as session events. Replay and observability may be reduced until the upstream channel is merged.
• LLM calls: enabling intent detection or using /init may send content to the configured auxiliary LLM and incur costs.
• Non-affiliation: this project is not affiliated with DeepSeek.

Contributing

Contributions are welcome. Before submitting a change, ensure:

npm run gate

passes completely.

When changing documentation, please update both README.md and README.zh-CN.md together.

License

Distributed under the MIT License.

Acknowledgments

• DeepSeek Harness
• Cordis
• Inspired by the context-economy design in docs/

This project is not affiliated with DeepSeek.