dsh-command-context-trim 0.1.0 → 0.1.1

package/CHANGELOG.md

@@ -5,6 +5,35 @@ All notable changes to this project are documented here. This project adheres to
 
 ## [Unreleased]
 
+## [0.1.1] - 2026-09-15
+
+### Fixed
+
+- **DeepSeek Harness 0.1.5 compatibility.** 0.1.5 renamed the positional replacement marker
+  (`{op: 'replace', start, end}` → `{op: 'replace', startSeq, endSeq}`), so every trim failed with
+  `session event "user/message" carries an invalid replace surfaceOp`. The plugin now probes the accepted shape
+  once against the harness actually installed and writes that one, so the same build works on the 0.1.2 and
+  0.1.5 lines without a version check.
+- **The system prompt is never trimmed — and no longer eats head protection.** 0.1.5 moved the system prompt
+  from the request header onto the surface as node 0 (`system/message`). A position-only "protect the first
+  node" rule would have protected the *system prompt* and exposed the **user's original request** as the first
+  elidable message. System nodes are now barriers: never elided and never crossed, and `protectHeadNodes` counts
+  only non-barrier nodes, so it keeps protecting the task statement.
+
+### Notes
+
+- No configuration changes. The only user-visible difference is the "fixed request overhead" refusal, which now
+  says "tool schemas and other non-surface request data": on 0.1.5 the system prompt is surface content rather
+  than header content.
+- Verified on harness 0.1.2-rc.1 and 0.1.5-rc.2 (40 tests each, same build).
+
+
+### Added
+
+- Integration tests against the **real** `ctx.tokenMeter` (bare cordis context + stub projection registry): a trim's
+  measured saving equals the `compaction/prune` shadow price it claims minus the replacement marker, and a *fresh* meter
+  folding the replayed log reaches the identical total — the replay property the claim exists for.
+
 ## [0.1.0] - 2026-09-15
 
 ### Added
@@ -32,5 +61,6 @@ All notable changes to this project are documented here. This project adheres to
   content stays in the durable session log. v1 has no `/untrim`.
 - Requires a harness that exposes `ctx.commands`, `ctx.tokenMeter`, and `ctx.llm` (DeepSeek Harness 0.1.2-rc.1 or later).
 
-[Unreleased]: https://github.com/snailium/dsh-command-context-trim/compare/v0.1.0...HEAD
+[Unreleased]: https://github.com/snailium/dsh-command-context-trim/compare/v0.1.1...HEAD
+[0.1.1]: https://github.com/snailium/dsh-command-context-trim/compare/v0.1.0...v0.1.1
 [0.1.0]: https://github.com/snailium/dsh-command-context-trim/releases/tag/v0.1.0

package/README.md

@@ -22,7 +22,7 @@ synchronous appends, zero LLM calls — it works precisely when every request is
 ## Install
 
 ```bash
-# from npm, once published
+# from npm
 dsh plugin --profile web add dsh-command-context-trim
 
 # from a checkout
@@ -78,6 +78,8 @@ Design consequences:
   original content stays in the durable session log, so a trim is auditable and recoverable by hand.
 - **No tool-call/result pair is ever split.** Cut edges are chosen with
   `toolPairingBalancedBefore`/`After` from `@deepseek-ai/dsh-compaction`.
+- **The system prompt is never trimmed** (harness 0.1.5+ carries it as a surface node): it is a barrier that no
+  elided span may touch or cross.
 - **Mutual exclusion with everything else.** The handler runs inside `agent.runMaintenance()`, which fails unless the agent
   is idle, so it cannot interleave with a turn, `/compact`, or automatic compaction; it also refuses while an unmatched
   `compaction/start` is open.
@@ -93,6 +95,26 @@ Design consequences:
 Within those bounds the policy is **oldest-first, least-long-possible**: the elided span starts at the oldest balanced cut
 and grows only until it frees exactly enough tokens.
 
+## Compatibility
+
+| Harness | State |
+|---|---|
+| 0.1.2-rc.1 (`latest`) | ✅ full suite green |
+| 0.1.5-rc.2 (`next`) | ✅ full suite green |
+
+Two harness changes between those lines are handled without a version check:
+
+- **The replacement marker was renamed** — `{op: 'replace', start, end}` became `{op: 'replace', startSeq, endSeq}`.
+  The plugin probes a throwaway detached session with each known shape at first use and writes the accepted one.
+- **The system prompt moved onto the surface** — 0.1.5 carries it as `system/message` node 0 instead of
+  `header.system`. System nodes are treated as **barriers**: they are never elided, no elided span crosses one,
+  and `protectHeadNodes` counts only non-barrier nodes, so head protection keeps covering the task statement
+  rather than the system prompt.
+
+Because of the second change, the fixed request overhead is now the tool schemas plus any other non-surface
+request data; on 0.1.2 it also included the system prompt. A trim's budget itself is unaffected — it comes from
+the token meter's total, whichever way the harness splits that total.
+
 ## Configuration
 
 Override on the `context-trim` row of a profile patch (the bundle's own `cordis.patch.yml` lists the full default set):
@@ -128,21 +150,26 @@ npm run link:harness   # or resolve @deepseek-ai from a local dsh installation i
 ```
 
 Tests cover the pure planner and argument parser, the surface mutation against a real `Session` (including log replay),
-and the plugin's command registration and end-to-end trim over a stub context. CI runs the suite on Node 22 and 24;
+the plugin's command registration and end-to-end trim over a stub context, and — against the real `ctx.tokenMeter` — that a
+trim's measured saving equals the shadow price it claims and that a fresh meter replaying the trimmed log lands on the very
+same total. CI runs the suite on Node 22 and 24;
 releases go out through `.github/workflows/publish.yml`, which is manual-only (`workflow_dispatch`).
 
 ### Verification status
 
 | Check | State |
 |---|---|
-| `npm test` (30 tests: planner, args, surface apply + replay, plugin handler) | ✅ passing |
+| `npm test` (34 tests: planner, args, surface apply + log replay, plugin handler) | ✅ passing |
 | Isolated `DSH_HOME` install (`dsh plugin add file:…`) reconciling dependency **and** bundle layer | ✅ verified |
 | Composed profile tree contains the `context-trim` insert row (`dsh --dump-config`) | ✅ verified |
 | Profile boot with the plugin mounted (no load error) | ✅ reaches the credential check cleanly |
-| Same suite against the pinned **published** harness packages (`npm ci`) | ✅ 30 passing |
-| CI workflow (Node 22 / 24) | ⏳ first run pending |
-| npm release via GitHub Actions | ⏳ first publish pending (Trusted Publishing cannot create a brand-new package; see the `npm-publish` procedure) |
-| End-to-end in the web GUI against a small-window model | ⏳ not yet run |
+| Same suite against the pinned **published** harness packages (`npm ci`) | ✅ 34 passing |
+| Integration against the **real** `ctx.tokenMeter`: measured drop equals the claimed shadow price, and a fresh meter replaying the trimmed log reaches the identical total | ✅ 4 tests |
+| Same suite on harness 0.1.5-rc.2 (renamed marker + surface system prompt) | ✅ 40 passing |
+| CI workflow (Node 22 / 24) | ✅ green |
+| npm release via GitHub Actions | ✅ 0.1.0 published with provenance (`+ dsh-command-context-trim@0.1.0`) |
+| Isolated profile install **from the npm registry** (dependency + bundle layer + composed insert row) | ✅ 0.1.0 |
+| End-to-end in the web GUI against a small-window model | ⏳ harness ready, not yet run |
 
 ## License
 

package/README.zh.md

@@ -17,7 +17,7 @@
 ## 安装
 
 ```bash
-dsh plugin --profile web add dsh-command-context-trim          # 发布到 npm 后
+dsh plugin --profile web add dsh-command-context-trim          # 从 npm 安装
 dsh plugin --profile web add file:/path/to/dsh-command-context-trim   # 从源码
 ```
 
@@ -58,6 +58,20 @@ dsh plugin --profile web add file:/path/to/dsh-command-context-trim   # 从源
 保护：开头 `protectHeadNodes`（默认 1，即任务声明）、末尾最近 `retainRatio` 窗口（下限 `minTailTokens`）、以及最后一条消息永不裁剪。
 在保护集之间采用**最旧优先、够用即止**：从最旧的平衡切点开始，只增长到刚好释放够 token。
 
+## 兼容性
+
+| Harness | 状态 |
+|---|---|
+| 0.1.2-rc.1（`latest`） | ✅ 全套测试通过 |
+| 0.1.5-rc.2（`next`） | ✅ 全套测试通过 |
+
+两处 0.1.5 变更已在不做版本号判断的前提下兼容：
+
+- **替换标记改名**：`{op:'replace', start, end}` → `{op:'replace', startSeq, endSeq}`。插件首次使用时用一个一次性游离 session 探测本机 harness 接受哪种形状，再按该形状写入。
+- **system prompt 从 header 搬到了 surface**（0.1.5 作为 `system/message` 节点 0）。system 节点被当作**屏障**：永不裁剪、任何被裁区间都不得跨越它；同时 `protectHeadNodes` 只统计非屏障节点，因此"保护头部"保护的仍是**用户的任务声明**，而不是 system prompt。
+
+因此"固定请求开销"的含义变为工具 schema + 其它非 surface 请求数据（0.1.2 上还包含 system prompt）；裁剪预算本身不受影响，因为它来自 token meter 的总量。
+
 ## 配置
 
 在 profile patch 的 `context-trim` 行上覆盖（`cordis.patch.yml` 里列出了全部默认值）：
@@ -82,10 +96,10 @@ npm test               # node --test
 npm run link:harness   # 也可改为从本地 dsh 安装的依赖闭包解析 @deepseek-ai
 ```
 
-已验证：30 个测试全部通过（选段算法、参数解析、真实 Session 上的 surface 改写与日志重放、插件命令注册与端到端裁剪）；
+已验证：34 个测试全部通过（选段算法、参数解析、真实 Session 上的 surface 改写与日志重放、插件命令注册与端到端裁剪，以及用**真实 `ctx.tokenMeter`** 验证「实测降幅 == 声明的 shadow price」和「新进程重放裁剪后日志得到完全一致的总量」）；
 隔离 `DSH_HOME` 安装后 dependency 与 bundle 层均正确 reconcile；`dsh --dump-config` 中出现 `context-trim` 行；profile 启动无加载错误。
 CI 在 Node 22/24 上跑同一套测试；发布通过 `.github/workflows/publish.yml`（手动 `workflow_dispatch`）。
-尚未执行：npm 首发（Trusted Publishing 无法创建全新包名，需要一次首发布引导）、以及 Web GUI 里的真实小窗口端到端验证。
+npm 0.1.1 已发布（带 provenance），并已在隔离 profile 里从 registry 安装验证；尚未执行：Web GUI 里的真实小窗口端到端验证（mock provider 与隔离实例已就绪）。
 
 ## License
 

package/lib/apply.js

@@ -21,6 +21,7 @@
  * @module dsh-command-context-trim/apply
  */
 import { createUserMessage } from '@deepseek-ai/dsh-llm';
+import { replacementOp } from './session-compat.js';
 
 /** Plugin name written into every replacement message's source marker. */
 export const TRIM_PLUGIN = 'dsh-command-context-trim';
@@ -81,17 +82,18 @@ export function isTrimMarkerSource(source) {
  * @param session - session whose surface is rewritten.
  * @param plan - committed span plan naming the shadowed range.
  * @param marker - replacement message built for that plan.
+ * @param keys - replacement-marker keys this harness accepts ({@link replaceKeys}).
  * @returns the appended replacement event.
  * @throws when the session rejects the append (surface contract violation).
  */
-export function applyTrim(session, plan, marker) {
+export function applyTrim(session, plan, marker, keys) {
 	session.append('compaction/prune', {
 		shadowedRange: { start: plan.startSeq, end: plan.endSeq },
 		shadowedSeqs: [...plan.shadowedSeqs],
 		shadowedTokenCount: plan.shadowedTokens
 	});
 	return session.append('user/message', marker, {
-		surfaceOp: { op: 'replace', start: plan.startSeq, end: plan.endSeq },
+		surfaceOp: replacementOp(keys, plan.startSeq, plan.endSeq),
 		sourceEventSeqs: [...plan.shadowedSeqs]
 	});
 }

package/lib/index.js

@@ -22,6 +22,7 @@ import { applyTrim, createMarkerMessage, provisionalMarker } from './apply.js';
 import { budgetFor, resolveConfig, retentionFor } from './config.js';
 import { planTrim } from './plan.js';
 import { resolveTarget, targetHeader } from './target.js';
+import { replaceKeys } from './session-compat.js';
 
 /** Cordis plugin name. */
 export const name = 'context-trim';
@@ -114,7 +115,15 @@ async function trimOnce(ctx, config, invocation, parsed, agentSignal) {
 	const label = target === undefined ? `an explicit ${budget}-token budget` : `${target.label} (window ${target.contextWindow})`;
 	const header = target === undefined ? undefined : targetHeader(session, target);
 	const measurement = ctx.tokenMeter.measure(session, header);
-	const nodes = measurement.nodes.map((node) => ({ seq: node.seq, heuristicTokens: node.heuristicTokens }));
+	const nodes = measurement.nodes.map((node) => ({
+		seq: node.seq,
+		heuristicTokens: node.heuristicTokens,
+		// Harness 0.1.5+ carries the system prompt as surface node 0. It is never
+		// elidable, and no elided span may cross it: dropping it would strip the
+		// model's instructions, and letting it consume head protection would expose
+		// the user's original request instead.
+		...(session.eventAt(node.seq)?.type === 'system/message' ? { barrier: true } : {})
+	}));
 	if (nodes.length === 0) {
 		return { kind: 'success', text: `Nothing to trim: ${session.id} has no model-visible messages yet.` };
 	}
@@ -148,7 +157,7 @@ async function trimOnce(ctx, config, invocation, parsed, agentSignal) {
 		markerTokens = finalMarkerTokens;
 	}
 	if (parsed.check) return { kind: 'success', text: preview(plan, label, markerTokens) };
-	const replacement = applyTrim(session, plan, marker);
+	const replacement = applyTrim(session, plan, marker, replaceKeys());
 	const after = ctx.tokenMeter.measure(session, header);
 	return { kind: 'success', text: report(plan, after, label, markerTokens), sourceEventSeq: replacement.seq };
 }
@@ -179,7 +188,7 @@ function describeNonSpan(plan, label) {
 			return {
 				kind: 'error',
 				text: [
-					`Fixed request overhead alone (~${plan.envelopeTokens} tokens of system prompt and tool schemas) exceeds the ${plan.budget}-token budget for ${label}.`,
+					`Fixed request overhead alone (~${plan.envelopeTokens} tokens of tool schemas and other non-surface request data) exceeds the ${plan.budget}-token budget for ${label}.`,
 					'Trimming conversation history cannot help: raise the backend context size (the model\'s contextWindow in settings.yaml, or the server\'s context flag) or reduce mounted tools and skills, then retry.'
 				].join('\n')
 			};
@@ -193,7 +202,7 @@ function describeNonSpan(plan, label) {
 				kind: 'error',
 				text: [
 					`Cannot free enough for ${label}: the largest balanced span frees ~${plan.maxFreeable} of the ~${plan.need} tokens needed.`,
-					`Protected content: task statement ~${plan.protectedHeadTokens} tokens, recent tail ~${plan.protectedTailTokens} tokens (retain target ~${plan.retainTokens}).`,
+					`Protected content: task statement ~${plan.protectedHeadTokens} tokens, recent tail ~${plan.protectedTailTokens} tokens (retain target ~${plan.retainTokens})${plan.hasBarrier ? ', plus the system prompt, which is never trimmed' : ''}.`,
 					'Try /compact (it summarizes instead of dropping), a larger window, or /trim with an explicit budget after another reduction.'
 				].join('\n')
 			};

package/lib/plan.js

@@ -8,6 +8,17 @@
  * and the original request are the two pieces of high-value context; everything
  * between them is what a smaller window can afford to lose.
  *
+ * Two node classes are never elidable:
+ *
+ * - **Barriers** (`barrier: true`) — the system prompt. Harness 0.1.5 moved it
+ *   from the request header onto the surface as node 0, which makes it
+ *   *trimmable* by position: dropping it would strip the model's instructions,
+ *   and letting it consume the head-protection budget would expose the user's
+ *   original request instead. Barriers are therefore untouchable and split the
+ *   elidable space into regions, and `protectHeadNodes` counts only non-barrier
+ *   nodes so it keeps protecting the task statement.
+ * - **The final node**, and the retained tail.
+ *
  * Nothing here touches a session: the planner receives measured node prices and
  * two balance predicates, so every branch is directly testable.
  *
@@ -18,17 +29,19 @@
  * @typedef {object} TrimNode
  * @property {number} seq - surface event sequence of the node.
  * @property {number} heuristicTokens - the token meter's heuristic price for it.
+ * @property {boolean} [barrier] - true for a node that may never be elided and
+ *   that no elided span may cross (the system prompt).
  */
 
 /**
  * @typedef {object} TrimPlanInput
  * @property {readonly TrimNode[]} nodes - current surface nodes in model-visible order.
- * @property {number} envelopeTokens - non-surface request price (system prompt + tool schemas).
+ * @property {number} envelopeTokens - non-surface request price (tool schemas and other fixed request data).
  * @property {number} budget - target total request size in tokens.
  * @property {number} markerCost - priced replacement marker, including configured slack.
  * @property {number} retainTokens - preferred verbatim recent-tail budget.
  * @property {number} minTailTokens - absolute floor for that retained tail.
- * @property {number} protectHeadNodes - leading nodes that must never be elided.
+ * @property {number} protectHeadNodes - leading non-barrier nodes that must never be elided.
  * @property {boolean} allowTailTrim - whether the elided span may reach into the retained tail.
  * @property {(seq: number) => boolean} isBalancedBefore - tool-pairing balance before a node.
  * @property {(seq: number) => boolean} isBalancedAfter - tool-pairing balance after a node.
@@ -45,7 +58,8 @@ export function planTrim(input) {
 		totalTokens: input.envelopeTokens + surfaceTokens,
 		surfaceTokens,
 		envelopeTokens: input.envelopeTokens,
-		budget: input.budget
+		budget: input.budget,
+		hasBarrier: input.nodes.some((node) => node.barrier === true)
 	};
 	if (common.totalTokens <= input.budget) return { kind: 'fits', ...common };
 	if (input.envelopeTokens >= input.budget) return { kind: 'envelope', ...common };
@@ -75,7 +89,7 @@ export function planTrim(input) {
 		...common,
 		need,
 		maxFreeable: weakest?.maxFreeable ?? 0,
-		protectedHeadTokens: headTokens(input.nodes, input.protectHeadNodes),
+		protectedHeadTokens: protectedHeadTokens(input.nodes, input.protectHeadNodes),
 		protectedTailTokens: weakest?.protectedTailTokens ?? 0,
 		retainTokens: configuredRetention
 	};
@@ -95,21 +109,39 @@ function attemptSpan(input) {
 		kind: 'shortfall',
 		maxFreeable,
 		protectedTailTokens: tailTokens(nodes, tailStart),
-		protectedHeadTokens: headTokens(nodes, input.protectHeadNodes)
+		protectedHeadTokens: protectedHeadTokens(nodes, input.protectHeadNodes)
 	});
-	const from = Math.min(input.protectHeadNodes, nodes.length - 1);
-	if (lastElidable < from) return shortfall(0);
+	const headEnd = protectedHeadEnd(nodes, input.protectHeadNodes);
+	if (lastElidable < headEnd) return shortfall(0);
+	let weakest = null;
+	for (const region of elidableRegions(nodes, headEnd, lastElidable)) {
+		const attempt = attemptRegion(nodes, region, input);
+		if (attempt === null) continue;
+		if (attempt.kind === 'span') return attempt;
+		if (weakest === null || attempt.maxFreeable > weakest.maxFreeable) weakest = attempt;
+	}
+	return shortfall(weakest?.maxFreeable ?? 0);
+}
+
+/**
+ * Grow the smallest balanced span from one region's oldest balanced cut.
+ * @param nodes - full priced surface.
+ * @param region - inclusive index range free of barriers.
+ * @param input - planning input carrying `need`, `markerCost`, and the balance predicates.
+ * @returns a `span` plan, or this region's best `shortfall`.
+ */
+function attemptRegion(nodes, region, input) {
 	let start = -1;
-	for (let index = from; index <= lastElidable; index += 1) {
+	for (let index = region.start; index <= region.end; index += 1) {
 		if (input.isBalancedBefore(nodes[index].seq)) {
 			start = index;
 			break;
 		}
 	}
-	if (start === -1) return shortfall(0);
+	if (start === -1) return null;
 	let accumulated = 0;
 	let best = null;
-	for (let index = start; index <= lastElidable; index += 1) {
+	for (let index = start; index <= region.end; index += 1) {
 		accumulated += nodes[index].heuristicTokens;
 		if (!input.isBalancedAfter(nodes[index].seq)) continue;
 		const freedTokens = accumulated - input.markerCost;
@@ -126,9 +158,34 @@ function attemptSpan(input) {
 		}
 		if (freedTokens >= input.need) break;
 	}
-	return best === null ? shortfall(0) : best.freedTokens >= input.need
-		? { kind: 'span', ...best }
-		: shortfall(best.freedTokens);
+	return best === null ? null : best.freedTokens >= input.need ? { kind: 'span', ...best } : { kind: 'shortfall', maxFreeable: best.freedTokens };
+}
+
+/** Index just past the `protectHeadNodes`-th non-barrier node. */
+function protectedHeadEnd(nodes, protectHeadNodes) {
+	let counted = 0;
+	for (let index = 0; index < nodes.length; index += 1) {
+		if (nodes[index].barrier === true) continue;
+		counted += 1;
+		if (counted >= protectHeadNodes) return index + 1;
+	}
+	return nodes.length;
+}
+
+/** Maximal runs of elidable indices inside `[from, to]`. */
+function elidableRegions(nodes, from, to) {
+	const regions = [];
+	let start = null;
+	for (let index = from; index <= to; index += 1) {
+		if (nodes[index].barrier === true) {
+			if (start !== null) regions.push({ start, end: index - 1 });
+			start = null;
+			continue;
+		}
+		if (start === null) start = index;
+	}
+	if (start !== null) regions.push({ start, end: to });
+	return regions;
 }
 
 /**
@@ -162,9 +219,9 @@ function tailTokens(nodes, tailStart) {
 	return total;
 }
 
-/** Tokens held by the protected leading nodes. */
-function headTokens(nodes, protectHeadNodes) {
+/** Tokens held by the protected prefix, barriers included. */
+function protectedHeadTokens(nodes, protectHeadNodes) {
 	let total = 0;
-	for (let index = 0; index < Math.min(protectHeadNodes, nodes.length); index += 1) total += nodes[index].heuristicTokens;
+	for (let index = 0; index < protectedHeadEnd(nodes, protectHeadNodes); index += 1) total += nodes[index].heuristicTokens;
 	return total;
 }

package/lib/session-compat.js

@@ -0,0 +1,80 @@
+/**
+ * Version tolerance for the harness's positional replacement marker.
+ *
+ * DeepSeek Harness renamed the surface-replacement keys in 0.1.5
+ * (`{op:'replace', start, end}` → `{op:'replace', startSeq, endSeq}`), and the
+ * session rejects any other shape at append time. Rather than parse a version
+ * string, the accepted shape is probed once against the harness actually
+ * installed: a throwaway detached session appends one replacement with each
+ * candidate and reports the first accepted.
+ *
+ * @module dsh-command-context-trim/session-compat
+ */
+import { createUserMessage } from '@deepseek-ai/dsh-llm';
+import { Session } from '@deepseek-ai/dsh-session';
+
+/** Candidate key sets, newest harness first. */
+const CANDIDATES = Object.freeze([
+	Object.freeze({ start: 'startSeq', end: 'endSeq' }),
+	Object.freeze({ start: 'start', end: 'end' })
+]);
+
+/** Cached probe result for the installed harness. */
+let cached;
+
+/**
+ * The replacement-marker keys this harness accepts.
+ * @returns `{ start, end }` key names.
+ * @throws when the harness accepts none of the known shapes.
+ */
+export function replaceKeys() {
+	cached ??= detectReplaceKeys();
+	return cached;
+}
+
+/**
+ * Probe every known marker shape against a detached session.
+ * @returns the first accepted key set.
+ * @throws when no known shape is accepted.
+ */
+export function detectReplaceKeys() {
+	for (const candidate of CANDIDATES) {
+		if (acceptsReplacement(candidate)) return candidate;
+	}
+	throw new Error(
+		'context-trim: this harness build accepts neither the 0.1.5+ nor the legacy positional replacement marker; ' +
+			'report it with the installed @deepseek-ai/dsh-session version'
+	);
+}
+
+/**
+ * Build the positional replacement marker in the harness's own key spelling.
+ * @param keys - key set returned by {@link replaceKeys}.
+ * @param startSeq - inclusive first shadowed surface seq.
+ * @param endSeq - inclusive last shadowed surface seq.
+ * @returns the marker to pass as `surfaceOp`.
+ */
+export function replacementOp(keys, startSeq, endSeq) {
+	return { op: 'replace', [keys.start]: startSeq, [keys.end]: endSeq };
+}
+
+/** Whether one detached session accepts a replacement written with these keys. */
+function acceptsReplacement(keys) {
+	try {
+		const session = Session.create('context-trim-probe');
+		session.append('user/message', probeMessage('probe first'), { surfaceOp: 'append' });
+		session.append('user/message', probeMessage('probe second'), { surfaceOp: 'append' });
+		session.append('user/message', probeMessage('probe marker'), {
+			surfaceOp: replacementOp(keys, 1, 1),
+			sourceEventSeqs: [1]
+		});
+		return true;
+	} catch {
+		return false;
+	}
+}
+
+/** One frozen probe message; detached sessions never publish it. */
+function probeMessage(text) {
+	return createUserMessage({ content: [{ type: 'text', text }], source: { kind: 'plugin', plugin: 'context-trim-probe' } });
+}

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "dsh-command-context-trim",
-  "version": "0.1.0",
+  "version": "0.1.1",
   "description": "Model-free /trim command for DeepSeek Harness — drop the oldest, least valuable span of conversation context so a session can continue on a smaller-window model, without any model call.",
   "license": "MIT",
   "author": "snailium",
@@ -35,14 +35,14 @@
     }
   },
   "peerDependencies": {
-    "@deepseek-ai/cordis": "^4.0.2",
-    "@deepseek-ai/dsh-commands": "^0.1.2-rc.1",
-    "@deepseek-ai/dsh-compaction": "^0.1.2-rc.1",
-    "@deepseek-ai/dsh-invariants": "^0.1.2-rc.1",
-    "@deepseek-ai/dsh-llm": "^0.1.2-rc.1",
-    "@deepseek-ai/dsh-session": "^0.1.2-rc.1",
-    "@deepseek-ai/dsh-token-meter": "^0.1.2-rc.1",
-    "@deepseek-ai/schemastery": "^3.18.1"
+    "@deepseek-ai/cordis": ">=4.0.2",
+    "@deepseek-ai/dsh-commands": ">=0.1.2-rc.1",
+    "@deepseek-ai/dsh-compaction": ">=0.1.2-rc.1",
+    "@deepseek-ai/dsh-invariants": ">=0.1.2-rc.1",
+    "@deepseek-ai/dsh-llm": ">=0.1.2-rc.1",
+    "@deepseek-ai/dsh-session": ">=0.1.2-rc.1",
+    "@deepseek-ai/dsh-token-meter": ">=0.1.2-rc.1",
+    "@deepseek-ai/schemastery": ">=3.18.1"
   },
   "peerDependenciesMeta": {
     "@deepseek-ai/cordis": {
@@ -71,9 +71,11 @@
     }
   },
   "devDependencies": {
+    "@deepseek-ai/cordis": "^4.0.2",
     "@deepseek-ai/dsh-compaction": "0.1.2-rc.1",
     "@deepseek-ai/dsh-llm": "0.1.2-rc.1",
     "@deepseek-ai/dsh-session": "0.1.2-rc.1",
+    "@deepseek-ai/dsh-token-meter": "^0.1.2-rc.1",
     "@deepseek-ai/schemastery": "3.18.2"
   },
   "keywords": [