dsh-asc
Agentic Surface Compaction (ASC) for DeepSeek Harness：模型决定何时以及压缩哪些内容，并将其提交为持久的会话日志替换项，支持完整重放、搜索和降级
插件会安装到这里；不确定时保持 web。
npx -y @deepseek-ai/dsh plugin --profile web add dsh-asc@0.2.0Asc 以 dsh-asc 发布，当前版本为 0.2.0。Plugin Hub 会校验它的 manifest，并保存精确安装来源，便于复现安装结果。
dsh-asc (full name DeepSeek Harness Agentic Surface Compaction) is a
context-compaction plugin for
DeepSeek Harness: the
model itself decides when and what to compact, and every compaction decision
is committed as a durable session-log replacement event
(surfaceOp: replace) — replayable, searchable, and reversible.
Inspired by the model-driven compaction philosophy of opencode-acp, but built on DSH's event-sourced log: compaction creates no side-state files, decompression is log replay, and search covers the full log including compacted originals.
Prerequisites: a working DeepSeek Harness
installation (dsh CLI available); Node.js ^22.19 or >=24.
From GitHub (the current distribution channel):
dsh plugin --profile <name> add github:lmst2/dsh-asc
dsh plugin adds the plugin to the profile and enables it automatically
based on the dsh.bundle declaration in the package; the tools and the
system prompt load together with that profile.
Restart required: after installing, restart the running DeepSeek Harness service.
From npm — once the package is published to the npm registry:
dsh plugin --profile <name> add dsh-asc
From source — to modify the plugin itself, or to contribute:
git clone https://github.com/lmst2/dsh-asc.git
cd dsh-asc
pnpm install
pnpm build
dsh plugin --profile <name> add "link:$(pwd)"
ctx.compaction allows only one provider at a time. Disable the default
basic backend in your profile's own cordis.patch.yml:
- id: compaction-basic
disabled: true
Optionally mount the invariant companion and the full-text-search backend:
- insert:
- id: dsh-asc-invariant # runtime invariant checks (optional, recommended)
name: "dsh-asc/invariant"
- id: session-query-sqlite # context_search full-text backend (optional)
name: "@deepseek-ai/dsh-session-query-sqlite"
After installing and restarting, no configuration is required — the plugin:
The plugin provides five model tools:
| Tool | Purpose |
|---|---|
context_status | context usage, tiered checkpoints, system/dialogue composition, recommended ranges, recent surface nodes |
context_compress | replace a surface range with a checkpoint you write (batching supported; tool-call pairs auto-extended; quality gate) |
context_decompress | undo a compaction: the original text returns to the surface at the checkpoint's own position (tier-aware; full: true reaches raw content) |
context_recap | re-read checkpoint summaries without decompressing the originals |
context_search | full-text search over the whole log (including compacted content) |
Compacted content is never lost: the originals stay in the session log and can be decompressed or searched at any time.
The system prompt ties the tools into one operating loop: capture consumed
raw work into tier-1 checkpoints, distill settled tier-1 piles into tier-2
decisions and tier-2 piles into a tier-3 fact index. Every checkpoint text
carries its topic and Compaction id, so when a visible summary already
points at the needed detail the model decompresses that block directly;
context_search is used only when no visible summary says where a detail
lives, and decompression always proceeds one tier at a time.
compaction/start → compaction/summary → replaced user/message →
compaction/end); no side state.src/
index.ts plugin entry: registers ctx.compaction + the five tools
config.ts strict config validation
types.ts shared config and result types
events.ts session-event vocabulary documentation (no custom members)
invariant.ts runtime invariant companion (subpath export)
engine/ the compaction engine core (engine, region, tier,
quality gate, fallback, prompt, restore)
policy/ protected-node policy and the nudge state machine
tools/ the five model tools
utils/ shared text helpers
tests/ vitest suites
docs/ usage, design, analysis, e2e-validation
| Doc | Contents |
|---|---|
| docs/usage.md | install, configuration, model experience, operations |
| docs/design.md | implemented contract: events, tools, automatic behavior, protection, invariants |
| docs/analysis.md | comparison of DSH and opencode-acp context management |
MIT. Algorithmic inspiration from DeepSeek Harness (MIT); only the ideas of opencode-acp (AGPL) are used, no source code. See NOTICE.