dsh-compact-after-task
A DSH (Cordis) plugin that auto-compacts a conversation after the current task finishes, whenever context pressure crosses a threshold you set (default 50% of the routed model's context window). Install it and stop manually clicking "compress" — and stop letting the context (and the bill) grow unchecked.
The switch and the threshold live on a dedicated card under Settings → Plugins → "Auto-compact" (toggle + slider); changes apply immediately, no restart needed.
The problem it solves
In a long conversation the model re-receives the entire history on every turn. As the conversation grows, this bites harder and harder:
- No compaction → exploding cost. The bigger the context, the more tokens each turn re-sends. Even with a provider KV cache (cheaper than fresh input, still not free), the history is re-transmitted and re-billed every single turn. Rolling from tens of thousands to hundreds of thousands of tokens quietly multiplies per-turn cost — the longer you talk, the more each reply costs, without you noticing.
- Manual compaction is tedious and easy to forget.
/compact or the UI compress button works, but you have to remember to use it — and people only remember once things are already slow and expensive, after several huge turns have been paid for.
- The built-in auto-compaction is a late, mid-task safety net.
dsh-compaction-basic only fires at 80% of the context window, and only at a step boundary while a task is running — by then you've already paid for several enormous requests, and the compaction interrupts work in progress.
- Slower responses. Longer prefixes mean more processing per request and a noticeably sluggish feel.
This plugin automates the "remember to compress" chore: after each task finishes (the agent returns to idle) it checks the pressure and compacts at your chosen threshold — dealing with cost before it hurts, not after.
How it relates to the built-in compaction (complementary, not a replacement)
| Built-in dsh-compaction-basic | This plugin |
|---|
| When | At each step boundary (agent/pre-step), mid-task | After the task fully finishes (agent returns to idle) |
| Threshold | thresholdRatio 0.8 (fixed) | thresholdRatio user-set, default 0.5 |
| Role | Last-resort mid-task safety net | Routine "slim down between tasks" |
The compaction itself is identical to /compact: it calls ctx.compaction.compactNow(), produces one <compacted-summary> checkpoint, and the conversation UI shows the usual "compacted N history items" card.
Requirements
- DSH Desktop / DSH profile with the base bundle (
dsh-base), which provides dsh-compaction-basic, dsh-token-meter, dsh-agent and dsh-llm.
- Core auto-compaction works on any version. The configuration form needs DSH ≥
0.1.7-alpha.2, because that release removed the third-party settings-card mechanism (Host-side settings.installSection, the settings.plugin.item slot and the settingsScope client service). This plugin instead declares its Config fields with .volatile(): the Plugins page auto-generates a form for the entry from that schema, keyed by the profile entry id (compact-after-task). A .volatile() field is a stable reference ({ get() }) at runtime — the loader commits a new immutable snapshot into the same reference when the user saves, so the plugin reads the current value on every use and needs neither a restart nor a change subscription. On runtimes without volatile support the same fields arrive as plain values, which the plugin also accepts.
- This plugin reaches the compaction service via
agentPresets.serviceFor(), not inject: ["compaction"]. Since agent presets arrived, the compaction service is no longer provided on the host plane — each session's preset (standard/cordis/…) mounts it inside an isolate realm, so a host-plane plugin that injects compaction pends forever (waiting for service: compaction). This plugin instead calls the official read API ctx.agentPresets.serviceFor(agent, "compaction") when the agent goes idle (that API exists precisely for callers whose request is about a session but arrives from outside it); on runtimes without agent presets it falls back to ctx.get("compaction").
Install
Two ways — pick one (do not mix them: the loader rejects a duplicate entry id).
A. Standard: plugin market / dsh plugin add (recommended)
The package ships a dsh.bundle.patch pointing at its own cordis.patch.yml, so it is a first-class bundle: installing it via the market or the CLI adds it to the profile's dsh.profile.bundles, and on the next boot the loader applies the package's own patch layer automatically — no manual editing needed.
# from npm (after publishing) — or from GitHub directly:
dsh plugin --profile desktop add @xiaobanli/dsh-compact-after-task
dsh plugin --profile desktop add github:xiaoliang2/dsh-compact-after-task
Or, in the desktop app, open Settings → Plugin Market and one-click install. Restart DSH Desktop, done.
B. Manual, profile-local (no market, no publishing)
-
Copy this package into the profile and make it resolvable from the profile's node_modules (same way dshmarket is installed):
# from the DSH home, e.g. C:\Users\you\.dsh\profiles\desktop
Copy-Item -Recurse <path-to>\dsh-compact-after-task .\plugins\dsh-compact-after-task
New-Item -ItemType Junction -Path .\node_modules\@xiaobanli\dsh-compact-after-task -Target .\plugins\dsh-compact-after-task
-
Add the plugin row to the profile's cordis.patch.yml (your user patch layer, applied after every bundle layer):
- insert:
- id: compact-after-task
name: '@xiaobanli/dsh-compact-after-task'
config:
enabled: true
thresholdRatio: 0.5
-
Declare the dependency in the profile's package.json so a later pnpm install does not prune it:
"dependencies": {
"@xiaobanli/dsh-compact-after-task": "file:plugins/dsh-compact-after-task"
}
-
Restart DSH Desktop for the profile change to take effect.
If you already installed it manually (method B), uninstall/remove it before installing via the market, otherwise the same entry id compact-after-task is inserted twice and the loader refuses to boot.
Configuration
| Key | Type | Default | Meaning |
|---|
enabled | boolean | true | Master switch. |
thresholdRatio | number (0.05–1) | 0.5 | Compress when pressure reaches this fraction of the routed model's context window (0.5 = 50%). |
onlyRoots | boolean | false | Only compress top-level conversation agents (skip background subagents). |
Adjusting the threshold
After install, open Settings → Plugins: beside the plugin list there is now an Auto-compact tab holding one card:
- Toggle — auto-compact after each task finishes;
- Slider — the threshold (5%–100%, 5% steps) with the live percentage beside it;
- Toggle — top-level conversations only (skip background subagents);
- Reset to default — shown only while the threshold is overridden, restoring the
cordis.patch.yml base-layer value.
The card binds its entry's settings namespace (compact-after-task) through the official client service ctx.configForms: saving writes into your profile's patch for that entry, and the loader commits the new value into the running plugin — no restart needed. (The same values are also editable from the plugin manager's auto-generated form for that row; the two are equivalent.)
Picking a threshold
Set it from "the max per-turn re-send you can tolerate ÷ the model's context window":
- DeepSeek's default 1M window: 0.3–0.6 is the sweet spot; 0.5 is a conservative default.
- Want each turn to re-send ≤ 400K tokens → 0.4; ≤ 250K → 0.25.
- Don't go ≥ 0.8: the built-in compaction already fires at 0.8 at a step boundary, so a higher threshold here is meaningless.
- Small-window models (128K/64K): raise to 0.6–0.8, or you'll squeeze out useful working context.
- Too low (below ~0.1) means compressing after every small task, and the summary swallows details that were still useful.
Tune by feel after a couple of long tasks: compressing after every little task / summaries getting coarse → raise it; still re-sending hundreds of thousands of tokens per turn → lower it. Drag the slider on the Auto-compact card (live) or edit cordis.patch.yml (base layer; restart).
Behavior details
- Listens to the
agent/status event and acts only on the idle transition — i.e. the current task's drive has fully drained. This is the "compress after finishing the current task" moment; it never interrupts a running task.
- Uses the same pressure measurement as the built-in engine:
tokenMeter.measure(session).totalTokens vs contextWindow × thresholdRatio (context window resolved via llm.resolveModelInfo).
- Compaction is a single
ctx.compaction.compactNow(agent, signal) call — the same operation /compact runs; compactNow reserves next-turn admission, so queued messages are correctly deferred until the compaction finishes. On rc.7+ ctx.compaction is the instance returned by agentPresets.serviceFor(agent, "compaction"); the operation is unchanged.
- Expected failures are silent:
busy (agent busy / a compaction already running) and cancelled are skipped; anything else is logged as a warning without disturbing agent lifecycle events.
- If the agent's preset mounts no compaction group (e.g.
minimal), serviceFor returns undefined and that agent is skipped.
Test
npm install
node test-compact-after-task.mjs
The unit test stubs the services and covers: idle + above threshold compacts, below threshold does not, running does not, busy is silent, disabled does nothing, onlyRoots filters subagents, the agentPresets.serviceFor and host-plane fallback paths, live .volatile() config reads (threshold and switch changed between events), and legacy plain-value config.
License
MIT
