Merge pull request #330 from deepseek-harness/compact-tool-pairing

compact: move tool-pairing helpers to compact seam
This commit is contained in:
Tianyi Cui
2026-07-17 22:03:07 +08:00
committed by GitHub
21 changed files with 518 additions and 395 deletions

View File

@@ -245,7 +245,7 @@ Creation announcement during session publication. A synchronous throw vetoes and
'session/created'(this: Scoped<Session>, session: Session): void
```
Source: [`packages/core/session/src/index.ts:47`](../../packages/core/session/src/index.ts)
Source: [`packages/core/session/src/index.ts:46`](../../packages/core/session/src/index.ts)
### `session/disposed` — emit
@@ -255,7 +255,7 @@ Emitted once when an announced session leaves the store, including publication r
'session/disposed'(this: Scoped<Session>, session: Session): void
```
Source: [`packages/core/session/src/index.ts:57`](../../packages/core/session/src/index.ts)
Source: [`packages/core/session/src/index.ts:56`](../../packages/core/session/src/index.ts)
### `session/event` — emit
@@ -267,7 +267,7 @@ Post-commit, fire-and-forget append feed. The listener snapshot resolves before
Types: [SessionEvent](../core-data-structures/core.md)
Source: [`packages/core/session/src/index.ts:69`](../../packages/core/session/src/index.ts)
Source: [`packages/core/session/src/index.ts:68`](../../packages/core/session/src/index.ts)
### `session/flush` — parallel
@@ -277,7 +277,7 @@ Awaited parallel durability checkpoint: every listener runs and the caller await
'session/flush'(this: Scoped<Session>, session: Session): Promise<void> | void
```
Source: [`packages/core/session/src/index.ts:79`](../../packages/core/session/src/index.ts)
Source: [`packages/core/session/src/index.ts:78`](../../packages/core/session/src/index.ts)
## `subagent/*`

View File

@@ -110,7 +110,7 @@ abstract compactRegion( session: Session, start: number, end: number, agent: Com
Types: [Message](../core-data-structures/core.md)
Source: [`packages/compact/compact/src/index.ts:36`](../../packages/compact/compact/src/index.ts)
Source: [`packages/compact/compact/src/index.ts:37`](../../packages/compact/compact/src/index.ts)
## `ctx.fs` — `FileSystem` (abstract seam)
@@ -219,7 +219,7 @@ list(): Session[]
fork(source: SessionForkSource, boundary?: number, childSessionId?: SessionId): Session
```
Source: [`packages/core/session/src/index.ts:540`](../../packages/core/session/src/index.ts)
Source: [`packages/core/session/src/index.ts:539`](../../packages/core/session/src/index.ts)
## `ctx.skills` — `SkillService`

View File

@@ -53,3 +53,5 @@ interface CompactionResult {
`CompactService` exposes `compactIfNeeded(...)` for pressure-triggered compaction, returning `null` when no compaction is needed, and `compactRegion(...)` for an explicit inclusive surface range. The pre-step caller supplies the agent, full prompt, session prefix, and abort signal; implementations must forward that signal to summarization. Estimation, retention, event sequencing, and summarization remain backend policy.
Auto-compaction runs at serial `agent/pre-step`, before the step and request derivation, so it can replace surface nodes while keeping trace events outside the step. Region boundaries preserve tool-call/result pairing but do not preserve whole turns, allowing early closed steps of one oversized turn to compact. `dsh-compact-basic` owns the retention and failure details.
The seam exports `toolPairingBalancedBefore(session, node)` and `toolPairingBalancedAfter(session, node)` for those edge checks. Both validate current surface membership, reject stale or missing seqs and orphan results, and ignore a caller-retained `node.next`; the [package contract](../../packages/compact/compact/README.md#tool-pairing-boundaries) owns their cache semantics.

View File

@@ -211,6 +211,8 @@ export interface SurfaceNode {
}
```
`SurfaceNode` is positional state, not durable identity. A replacement can remove a caller-retained node or make a copied `next` stale; consumers that cross a surface mutation validate membership and answer positional queries from `Session.surface.nodes`. `SurfaceManager.replaceGeneration` increments for each replacement so incremental consumers can distinguish pure tail growth from a rewrite.
### `SurfaceFoldReplacement` and `SurfaceFoldResult` — a complete surface replay
`foldSurface(events)` returns detached current nodes together with the actual node seqs shadowed by each declared replacement range. `SurfaceManager` uses the same transition functions for its incremental cache.

View File

@@ -25,10 +25,10 @@ This matrix shows which packages dispatch each harness-owned event and which pac
| `fs/observed` | `emit` | [`packages/fs/fs/src/index.ts:70`](../packages/fs/fs/src/index.ts) | [`tool-fs`](../packages/fs/tool-fs) (`emit`) | [`fs-policy`](../packages/fs/fs-policy) |
| `fs/write-intent` | `waterfall` | [`packages/fs/fs/src/index.ts:53`](../packages/fs/fs/src/index.ts) | [`tool-fs`](../packages/fs/tool-fs) (`waterfall`) | [`fs-policy`](../packages/fs/fs-policy) |
| `llm/stream` | `waterfall` | [`packages/llm/llm/src/index.ts:39`](../packages/llm/llm/src/index.ts) | [`llm`](../packages/llm/llm) (`waterfall`) | [`invariants`](../packages/support/invariants), [`llm-replay`](../packages/support/llm-replay) |
| `session/created` | `emit` | [`packages/core/session/src/index.ts:47`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`invariants`](../packages/support/invariants), [`jsonrpc`](../packages/ui/jsonrpc), [`session-persistence`](../packages/session-persistence/session-persistence) |
| `session/disposed` | `emit` | [`packages/core/session/src/index.ts:57`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`session-persistence`](../packages/session-persistence/session-persistence) |
| `session/event` | `emit` | [`packages/core/session/src/index.ts:69`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`acp`](../packages/ui/acp), [`invariants`](../packages/support/invariants), [`jsonrpc`](../packages/ui/jsonrpc), [`session-persistence`](../packages/session-persistence/session-persistence), [`stdio`](../packages/ui/stdio), [`workspace-context`](../packages/context/workspace-context) |
| `session/flush` | `parallel` | [`packages/core/session/src/index.ts:79`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`session-persistence`](../packages/session-persistence/session-persistence) |
| `session/created` | `emit` | [`packages/core/session/src/index.ts:46`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`invariants`](../packages/support/invariants), [`jsonrpc`](../packages/ui/jsonrpc), [`session-persistence`](../packages/session-persistence/session-persistence) |
| `session/disposed` | `emit` | [`packages/core/session/src/index.ts:56`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`session-persistence`](../packages/session-persistence/session-persistence) |
| `session/event` | `emit` | [`packages/core/session/src/index.ts:68`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`acp`](../packages/ui/acp), [`invariants`](../packages/support/invariants), [`jsonrpc`](../packages/ui/jsonrpc), [`session-persistence`](../packages/session-persistence/session-persistence), [`stdio`](../packages/ui/stdio), [`workspace-context`](../packages/context/workspace-context) |
| `session/flush` | `parallel` | [`packages/core/session/src/index.ts:78`](../packages/core/session/src/index.ts) | [`session`](../packages/core/session) (`events.dispatch`) | [`session-persistence`](../packages/session-persistence/session-persistence) |
| `subagent/end` | `emit` | [`packages/subagent/subagent/src/index.ts:108`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`events.dispatch`) | [`hooks-claude`](../packages/hooks/hooks-claude), [`jsonrpc`](../packages/ui/jsonrpc) |
| `subagent/provider-added` | `emit` | [`packages/subagent/subagent/src/index.ts:82`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`emit`) | [`tool-subagent`](../packages/subagent/tool-subagent) |
| `subagent/provider-removed` | `emit` | [`packages/subagent/subagent/src/index.ts:88`](../packages/subagent/subagent/src/index.ts) | [`subagent`](../packages/subagent/subagent) (`events.dispatch`) | [`tool-subagent`](../packages/subagent/tool-subagent) |

View File

@@ -52,7 +52,7 @@ The loop derives messages once after `agent/pre-step`. Running before `step/star
Auto-compaction fires before **every** step, not once per turn. This is **load-bearing for runaway-turn survival**: a tool-heavy ReAct turn appends an `assistant/message` + a `tool/result` per step, so the surface grows *within* a turn. A single turn can grow past the window on its own (a "runaway turn") — and the only moment to rescue it before the next model call overflows is the next step's `pre-step` checkpoint. Gating compaction to a turn's first step (or, worse, retaining the whole in-flight turn verbatim) re-opens exactly the hole compaction exists to close: the harness would die when compaction is most needed.
`compactIfNeeded` retains the smallest tail of whole surface units whose estimated size reaches `retainTokens` and compacts older nodes. A unit is a complete closed step or one no-step message. If the token cutoff lands inside a step, retention expands until the cut is tool-pairing balanced. Balance is checked on surface order, not log sequence, because replacement summaries have new sequence numbers at old surface positions. `compactRegion` rejects boundaries that split a tool call from its result. The in-flight turn receives no special retention.
`compactIfNeeded` retains the smallest tail of whole surface units whose estimated size reaches `retainTokens` and compacts older nodes. A unit is a complete closed step or one no-step message. If the token cutoff lands inside a step, retention expands until the cut is tool-pairing balanced. Balance is checked on surface order, not log sequence, because replacement summaries have new sequence numbers at old surface positions. `dsh-compact` exports the before/after edge helpers; their per-session cache folds only appended surface-tail nodes while `replaceGeneration` is unchanged, does no event reads for log-only growth, and rebuilds current membership and balances after replacement. `compactRegion` rejects boundaries that split a tool call from its result. The in-flight turn receives no special retention.
A runaway turn thus compacts exactly like any other history: its early *closed* steps get summarized while its recent steps stay verbatim. When the only compactable content left is an un-splittable open tail step (its tool-calls have no results yet), compaction declines (`null`) and retries once that step closes.
@@ -113,7 +113,7 @@ Two failure paths, both documented:
- **New packages**: `packages/compact/compact` (interface) and a sibling `compact-basic` (backend) under `packages/compact/`, wired into the root tsconfigs. The consumer tier is deferred.
- **New loop seam**: `agent/pre-step` (`@mode serial`) declared in `dsh-agent` and emitted by `dsh-agent-loop` after system assembly and before `step/start`. This is a documented change to the loop — `docs/architecture.md` records it and the generated cordis catalog carries its signature.
- **`SessionEventMap`** gains `compact/start` / `compact/summary` / `compact/end` by declaration merging (merge-extensible); `SurfaceEventType` is **not** touched. These are session events, not cordis `Events`, so the event-taxonomy gate needs no entry.
- **`dsh-session`** gains the tool-pairing balance predicate (`isToolPairingBalanced`, in `tool-pairing.ts`, exported from the package index) that `compactRegion`/`compactIfNeeded` use to keep a collapsed region from splitting a step's tool-call/result pair. The surface `replace` op and the surface-metadata runtime guard already existed and are reused.
- **`dsh-compact`** owns `toolPairingBalancedBefore(session, node)` and `toolPairingBalancedAfter(session, node)`, the cached surface-edge checks that `compactRegion` and `compactIfNeeded` use to avoid splitting a tool-call/result pair. The cache validates current membership by seq and answers both edges from one per-cut balance sequence instead of trusting a caller-retained `node.next`; stale or missing seqs and orphan results reject. `dsh-session` continues to own the surface `replace` operation, positional nodes, and rewrite generation.
- **`dsh-invariants`** drops its `surface replace: start must be <= end` assertion: a head-anchored compaction lands a high-seq replacement node at an older range's *position*, so `start > end` numerically is normal and valid (the range is positional, validated by the surface's `indexOf` checks that remain). The turn-enclosure invariant is reused unchanged.
- **Wiring**: `dsh-compact-basic` is loaded in `examples/coding-agent`'s `cordis.yml`, so the seam ships in the real demo (it was previously loaded nowhere).

View File

@@ -6,7 +6,7 @@ Status: proposed
The session log maintains two representations that cost more machinery than their consumers require: a pseudo-linked surface and custom request-header deltas.
`SurfaceManager` stores the same order in an array, a seq map, and mutable `prev`/`next` links. Production never reads `prev`; compact's sole `next` read is the successor of an array position. Replacement already uses `indexOf`, so the links do not make its dominant operation constant-time. A seq array with linear replacement lookup has the same asymptotic replacement cost and one representation to validate.
`SurfaceManager` stores the same order in an array, a seq map, and mutable `prev`/`next` links. Production never reads either link: compact's tool-pairing balance answers from per-cut balances cached in surface order. Replacement already uses `indexOf`, so the links do not make its dominant operation constant-time. A seq array with linear replacement lookup has the same asymptotic replacement cost and one representation to validate.
The request-header subsystem implements a custom system/tool delta codec and transmission-decision layer even though its contract says deltas are an encoding optimization, not a reconstructability requirement. Retaining the initial/resume full snapshot at each loop-instance boundary, then writing a canonical full `request/header` whenever that instance's assembled header changes, preserves replay while deleting `SystemDelta`, `ToolsDelta`, round-trip fallback, and the durable `request/header-delta` variant. Codec-only vocabulary disappears with the codec, not because its individual arms were invalid.

View File

@@ -9,7 +9,7 @@ This is the implementation tier of the compaction capability — see the [interf
This backend owns the compaction policy:
- **Estimation** — a configurable characters-per-token heuristic counts the current session prefix supplied to pre-step, derived history, and system prompt, matching the next request rather than stale logged prefix state.
- **Retention** — compact the oldest whole surface units while preserving a recent tail and balanced tool-call/result cuts. Turn boundaries do not protect old steps inside a runaway turn. An open indivisible tail declines until it closes; a single unit larger than the budget remains out of scope.
- **Retention** — compact the oldest whole surface units while preserving a recent tail and balanced tool-call/result cuts through the [`dsh-compact` boundary helpers](../compact/README.md#tool-pairing-boundaries). Turn boundaries do not protect old steps inside a runaway turn. An open indivisible tail declines until it closes; a single unit larger than the budget remains out of scope.
- **Convergence** — retry head-checkpoint compaction up to `compactionRetries`; reject a summary that does not shrink its source, and throw if retries cannot return below threshold.
- **Summarization** — a direct `llm/stream` call uses the configured model and cap without running the loop-only `agent/request` seam. The input transcript preserves non-text blocks as tagged placeholders; only returned text enters the checkpoint, excluding reasoning and tool calls that would leak private reasoning or create an orphaned call.
- **Framing** — the replacement user message marks established checkpoint context with `<compacted-summary>` tags. The raw summary remains on the provenance event, and later automatic cycles merge the prior checkpoint.

View File

@@ -7,12 +7,11 @@
*/
import { Context } from 'cordis'
import { CompactService, renderTranscript } from '@deepseek-ai/dsh-compact'
import { CompactService, renderTranscript, toolPairingBalancedAfter, toolPairingBalancedBefore } from '@deepseek-ai/dsh-compact'
import type { CompactionResult } from '@deepseek-ai/dsh-compact'
import { BlockAssembler } from '@deepseek-ai/dsh-llm'
import type { ContentBlock, FinishReason, GenerateOptions, Message } from '@deepseek-ai/dsh-llm'
import type { Session, SessionEvent } from '@deepseek-ai/dsh-session'
import { isToolPairingBalanced } from '@deepseek-ai/dsh-session'
import type { Agent } from '@deepseek-ai/dsh-agent'
import type { BasicCompactConfig, ResolvedConfig } from './types.ts'
import { resolveConfig } from './types.ts'
@@ -348,15 +347,14 @@ export class BasicCompactService extends CompactService {
}
// Both range edges must preserve assistant tool-call/result pairing.
const events = session.events
if (!isToolPairingBalanced(nodes, events, start)) {
// eslint-disable-next-line @typescript-eslint/no-non-null-assertion
const startNode = nodes[startIdx]!
if (!toolPairingBalancedBefore(session, startNode)) {
throw new Error(`compactRegion: start seq ${start} is not a balanced boundary (would split a step's tool-call/result pair)`)
}
// The cut after `end` is named by `end`'s surface successor, or `null` when
// `end` is the tail.
// eslint-disable-next-line @typescript-eslint/no-non-null-assertion
const afterEnd: number | null = nodes[endIdx]!.next
if (!isToolPairingBalanced(nodes, events, afterEnd)) {
const endNode = nodes[endIdx]!
if (!toolPairingBalancedAfter(session, endNode)) {
throw new Error(`compactRegion: end seq ${end} is not a balanced boundary (would split a step, or the step is still open)`)
}
@@ -511,7 +509,7 @@ export class BasicCompactService extends CompactService {
// splitting an assistant↔result pair.
while (keepFromIdx > 0) {
// eslint-disable-next-line @typescript-eslint/no-non-null-assertion
if (isToolPairingBalanced(nodes, events, nodes[keepFromIdx]!.seq)) break
if (toolPairingBalancedBefore(session, nodes[keepFromIdx]!)) break
keepFromIdx -= 1
}
if (keepFromIdx === 0) return null

View File

@@ -1,8 +1,8 @@
import { describe, expect, it } from 'vitest'
import { Context } from 'cordis'
import { toolPairingBalancedAfter, toolPairingBalancedBefore } from '@deepseek-ai/dsh-compact'
import type { ContentBlock, GenerateOptions, StreamChunk } from '@deepseek-ai/dsh-llm'
import { CallId, LlmAdapter } from '@deepseek-ai/dsh-llm'
import { isToolPairingBalanced } from '@deepseek-ai/dsh-session'
import { defineTool } from '@deepseek-ai/dsh-tools'
import { AgentId } from '@deepseek-ai/dsh-agent'
import AgentLoop, { ReactLoopAgent } from '@deepseek-ai/dsh-agent-loop'
@@ -118,9 +118,9 @@ describe('CBR-001: a real-loop checkpoint is a valid boundary on both sides', ()
for (const cp of checkpoints) {
const node = nodes.find(n => n.seq === cp.seq)
if (!node) continue // shadowed by a later checkpoint — no longer an edge.
expect(isToolPairingBalanced(nodes, events, node.seq),
expect(toolPairingBalancedBefore(agent.session, node),
`checkpoint seq ${node.seq} must be a balanced region START`).toBe(true)
expect(isToolPairingBalanced(nodes, events, node.next),
expect(toolPairingBalancedAfter(agent.session, node),
`checkpoint seq ${node.seq} must be a balanced region END`).toBe(true)
}
} finally {

View File

@@ -6,7 +6,7 @@ This package is the interface tier of the compaction capability, split so each c
| Package | Role |
|---|---|
| `@deepseek-ai/dsh-compact` (this) | the interface: abstract service + `compact/*` events + `CompactionResult` + the shared transcript renderer (`renderTranscript`/`renderContentBlocks`) |
| `@deepseek-ai/dsh-compact` (this) | the interface: abstract service + `compact/*` events + `CompactionResult` + tool-pairing boundary helpers + the shared transcript renderer (`renderTranscript`/`renderContentBlocks`) |
| `@deepseek-ai/dsh-compact-basic` | a backend: chars-per-token estimation (`charsPerToken`, default 4) + token-budget retention + `llm.stream()` summarization |
| `@deepseek-ai/dsh-tool-compact` (deferred) | the model-facing `/compact` tool over `ctx.compact` |
@@ -23,6 +23,12 @@ Both methods are **abstract** — the backend owns the entire strategy (token es
`compactIfNeeded` takes a required `signal`; `compactRegion`'s is optional. A backend that summarizes via `ctx.llm.stream()` **must** forward it into the call's `GenerateOptions.signal`, so an abort or fiber dispose tears down the in-flight summarization instead of leaving an orphaned model call running past the cancellation. The session being compacted comes from the agent context; the turn that the `compact/*` events belong to is recoverable from the log (the currently-open turn), so the backend stamps it from the log rather than trusting a caller-supplied value.
## Tool-pairing boundaries
The interface exports `toolPairingBalancedBefore(session, node)` and `toolPairingBalancedAfter(session, node)` for snapping and validating compaction edges. A safe edge has no unanswered assistant tool call crossing it. Each helper identifies the node by seq alone and answers from balances cached per cut in current surface order, so a stale caller-held `node.next` cannot choose the cut.
The private per-session cache is keyed by `session.surface.replaceGeneration` and the processed surface-node count. An unchanged generation extends the fold with unseen tail nodes only; a log-only append with no new surface node does no event reads, while a replacement generation rebuilds current membership and balances. Missing event seqs and a `tool/result` without a preceding open call reject as corrupt surface state.
## Surface contract
`SurfaceEventType` is a closed union — only `user/message`, `assistant/message`, `tool/result`, `context/message`, and `steering/message` may carry `surfaceOp`. A `compact/*` event therefore **cannot** appear on the surface. A successful compaction instead:

View File

@@ -14,6 +14,7 @@ import type { CompactionResult } from './types.ts'
export type { CompactionResult } from './types.ts'
export { renderContentBlocks, renderTranscript } from './render.ts'
export { toolPairingBalancedAfter, toolPairingBalancedBefore } from './tool-pairing.ts'
/** Minimal agent context compaction needs without depending on the agent package. */
export interface CompactAgentContext {
@@ -67,6 +68,8 @@ export abstract class CompactService extends Service {
* balanced so assistant tool calls remain paired with their results. A model-
* backed implementation forwards cancellation and rejects active, missing,
* reversed, or unbalanced ranges.
* Use {@link toolPairingBalancedBefore} and {@link toolPairingBalancedAfter}
* for the edge checks.
*
* @param session - session to mutate.
* @param start - first surface seq, inclusive.

View File

@@ -0,0 +1,132 @@
/**
* Tool-pairing balance over a session surface. Compaction changes surface
* positions, so safe cuts are derived from tool-call/result content in current
* surface order rather than step markers or linked-list fields supplied by a
* caller.
* @module @deepseek-ai/dsh-compact/tool-pairing
*/
import type { Session, SessionEvent, SurfaceNode } from '@deepseek-ai/dsh-session'
/** Incremental balance state for one session surface generation. */
interface BalanceCache {
/** Surface rewrite generation this state describes. */
generation: number
/**
* Balance of every surface cut in current order: a surface of N nodes has
* N + 1 cuts, entry `i` being the cut before node `i` and the final entry
* the cut after the surface tail.
*/
cutBalanced: readonly boolean[]
/** Current surface position of each node seq, indexing {@link cutBalanced}. */
indexBySeq: Map<number, number>
/** In-progress tool-call count after the processed surface tail. */
inProgressToolCalls: number
}
const balanceCacheBySession = new WeakMap<Session, BalanceCache>()
/** Return how one surface event changes the in-progress tool-call count. */
function nodeDelta(event: SessionEvent): number {
switch (event.type) {
case 'assistant/message':
return event.data.content.filter(block => block.type === 'tool-call').length
case 'tool/result':
return -1
default:
return 0
}
}
/** Read and validate the event named by a surface node. */
function eventForNode(events: readonly SessionEvent[], node: SurfaceNode): SessionEvent {
const event = events[node.seq]
if (event === undefined || event.seq !== node.seq) {
throw new Error(`tool-pairing balance: surface seq ${node.seq} has no matching session event (corrupt surface)`)
}
return event
}
/** Fold surface nodes not yet in the cache into its balance state. */
function extendCache(
session: Session,
cache: BalanceCache,
nodes: readonly SurfaceNode[],
): BalanceCache {
const processed = cache.cutBalanced.length - 1
const tail = nodes.slice(processed)
// Validate the unseen tail before mutating the live cache, so a corrupt
// append cannot leave a partially advanced state behind.
const events = session.events
const pendingCuts: boolean[] = []
let inProgressToolCalls = cache.inProgressToolCalls
for (const node of tail) {
inProgressToolCalls += nodeDelta(eventForNode(events, node))
if (inProgressToolCalls < 0) {
throw new Error(`tool-pairing balance: tool/result at surface seq ${node.seq} has no matching tool-call (corrupt surface)`)
}
pendingCuts.push(inProgressToolCalls === 0)
}
tail.forEach((node, offset) => cache.indexBySeq.set(node.seq, processed + offset))
cache.cutBalanced = cache.cutBalanced.concat(pendingCuts)
cache.inProgressToolCalls = inProgressToolCalls
return cache
}
/** Return balance state synchronized with the current session surface. */
function balanceCache(session: Session): BalanceCache {
const surface = session.surface
const nodes = surface.nodes
const generation = surface.replaceGeneration
const cached = balanceCacheBySession.get(session)
if (cached === undefined || cached.generation !== generation || cached.cutBalanced.length - 1 > nodes.length) {
// A rebuild is the same fold started from the empty-surface state, whose
// single leading cut is trivially balanced.
const rebuilt = extendCache(session, {
generation,
cutBalanced: [true],
indexBySeq: new Map(),
inProgressToolCalls: 0,
}, nodes)
balanceCacheBySession.set(session, rebuilt)
return rebuilt
}
if (cached.cutBalanced.length - 1 < nodes.length) return extendCache(session, cached, nodes)
return cached
}
/** Balance of the cut at a node's position plus offset, rejecting seqs outside current membership. */
function cutBalance(cache: BalanceCache, seq: number, offset: 0 | 1): boolean {
const index = cache.indexBySeq.get(seq)
const balanced = index === undefined ? undefined : cache.cutBalanced[index + offset]
if (balanced === undefined) {
throw new Error(`tool-pairing balance: surface seq ${seq} not found`)
}
return balanced
}
/**
* Whether the cut immediately before a current surface node is tool-pairing balanced.
* @param session - session whose surface is checked.
* @param node - surface node whose leading cut is checked; only its seq identifies it.
* @returns true when no unanswered tool call crosses the cut.
* @throws when the seq is absent from the current surface, a surface node has no
* matching log event, or a tool result has no preceding open call.
*/
export function toolPairingBalancedBefore(session: Session, node: SurfaceNode): boolean {
return cutBalance(balanceCache(session), node.seq, 0)
}
/**
* Whether the cut immediately after a current surface node is tool-pairing balanced.
* @param session - session whose surface is checked.
* @param node - surface node whose trailing cut is checked; only its seq identifies it.
* @returns true when no unanswered tool call crosses the cut.
* @throws when the seq is absent from the current surface, a surface node has no
* matching log event, or a tool result has no preceding open call.
*/
export function toolPairingBalancedAfter(session: Session, node: SurfaceNode): boolean {
return cutBalance(balanceCache(session), node.seq, 1)
}

View File

@@ -0,0 +1,330 @@
import { describe, expect, it } from 'vitest'
import { CallId } from '@deepseek-ai/dsh-llm'
import { toolPairingBalancedAfter, toolPairingBalancedBefore } from '@deepseek-ai/dsh-compact'
import { Session, SessionId } from '@deepseek-ai/dsh-session'
import type { SessionEvent, SurfaceNode } from '@deepseek-ai/dsh-session'
const SURFACE = { surfaceOp: 'append' as const }
function seqOf(session: Session, type: SessionEvent['type'], nth = 0): number {
return session.events.filter(event => event.type === type)[nth]!.seq
}
function nodeAt(session: Session, seq: number): SurfaceNode {
const node = session.surface.nodes.find(candidate => candidate.seq === seq)
if (node === undefined) throw new Error(`seq ${seq} is not a surface node`)
return node
}
function before(session: Session, type: SessionEvent['type'], nth = 0): boolean {
return toolPairingBalancedBefore(session, nodeAt(session, seqOf(session, type, nth)))
}
function after(session: Session, type: SessionEvent['type'], nth = 0): boolean {
return toolPairingBalancedAfter(session, nodeAt(session, seqOf(session, type, nth)))
}
function closedToolStep(): Session {
const session = new Session(SessionId('closed-tool-step'))
session.append('user/message', {
content: [{ type: 'text', text: 'go' }],
source: { kind: 'user' },
}, SURFACE)
session.append('assistant/message', {
turn: 1,
step: 1,
content: [{ type: 'tool-call', id: CallId('c1'), name: 'bash', arguments: '{}' }],
}, SURFACE)
session.append('tool/result', {
turn: 1,
step: 1,
callId: CallId('c1'),
content: [{ type: 'text', text: 'done' }],
isError: false,
}, SURFACE)
return session
}
describe('tool-pairing boundaries', () => {
it('classifies closed and open single-call steps', () => {
const closed = closedToolStep()
expect(before(closed, 'user/message')).toBe(true)
expect(after(closed, 'user/message')).toBe(true)
expect(before(closed, 'assistant/message')).toBe(true)
expect(after(closed, 'assistant/message')).toBe(false)
expect(before(closed, 'tool/result')).toBe(false)
expect(after(closed, 'tool/result')).toBe(true)
const open = new Session(SessionId('open-tool-step'))
open.append('assistant/message', {
turn: 1,
step: 1,
content: [{ type: 'tool-call', id: CallId('open'), name: 'bash', arguments: '{}' }],
}, SURFACE)
expect(toolPairingBalancedAfter(open, open.surface.nodes[0]!)).toBe(false)
})
it('requires every result from a multiple-call assistant message', () => {
const session = new Session(SessionId('multiple-calls'))
session.append('assistant/message', {
turn: 1,
step: 1,
content: [
{ type: 'tool-call', id: CallId('c1'), name: 'one', arguments: '{}' },
{ type: 'tool-call', id: CallId('c2'), name: 'two', arguments: '{}' },
],
}, SURFACE)
session.append('tool/result', {
turn: 1, step: 1, callId: CallId('c1'), content: [], isError: false,
}, SURFACE)
session.append('tool/result', {
turn: 1, step: 1, callId: CallId('c2'), content: [], isError: false,
}, SURFACE)
expect(after(session, 'tool/result', 0)).toBe(false)
expect(after(session, 'tool/result', 1)).toBe(true)
})
it('keeps neutral nodes inside an open pair unbalanced and free nodes balanced', () => {
const midStep = new Session(SessionId('neutral-mid-step'))
midStep.append('assistant/message', {
turn: 1,
step: 1,
content: [{ type: 'tool-call', id: CallId('c1'), name: 'bash', arguments: '{}' }],
}, SURFACE)
midStep.append('context/message', {
content: [{ type: 'text', text: 'background update' }],
source: { kind: 'plugin', plugin: 'test' },
}, SURFACE)
midStep.append('tool/result', {
turn: 1, step: 1, callId: CallId('c1'), content: [], isError: false,
}, SURFACE)
expect(before(midStep, 'context/message')).toBe(false)
expect(after(midStep, 'context/message')).toBe(false)
const free = new Session(SessionId('neutral-free'))
free.append('context/message', {
content: [{ type: 'text', text: 'idle injection' }],
source: { kind: 'user' },
}, SURFACE)
expect(before(free, 'context/message')).toBe(true)
expect(after(free, 'context/message')).toBe(true)
})
})
describe('tool-pairing surface identity', () => {
it('rebuilds after replace and rejects nodes removed from current membership', () => {
const session = closedToolStep()
const staleTail = nodeAt(session, seqOf(session, 'tool/result'))
expect(toolPairingBalancedAfter(session, staleTail)).toBe(true)
const nodes = session.surface.nodes
session.append('user/message', {
content: [{ type: 'text', text: 'checkpoint' }],
source: { kind: 'plugin', plugin: 'compact' },
}, {
surfaceOp: { op: 'replace', start: nodes[0]!.seq, end: nodes.at(-1)!.seq },
sourceEventSeqs: nodes.map(node => node.seq),
})
const checkpoint = session.surface.nodes[0]!
expect(toolPairingBalancedBefore(session, checkpoint)).toBe(true)
expect(toolPairingBalancedAfter(session, checkpoint)).toBe(true)
expect(() => toolPairingBalancedBefore(session, staleTail)).toThrow(/surface seq .* not found/)
expect(() => toolPairingBalancedAfter(session, staleTail)).toThrow(/surface seq .* not found/)
})
it('ignores a caller-held node next field and answers from cached balances', () => {
const session = closedToolStep()
const assistant = nodeAt(session, seqOf(session, 'assistant/message'))
expect(toolPairingBalancedAfter(session, { ...assistant, next: null })).toBe(false)
expect(toolPairingBalancedAfter(session, { ...assistant, next: 999 })).toBe(false)
})
it('rejects missing seqs before and after, including an empty surface', () => {
const session = new Session(SessionId('missing-membership'))
const missing: SurfaceNode = { seq: 999, prev: null, next: null }
expect(() => toolPairingBalancedBefore(session, missing)).toThrow(/surface seq 999 not found/)
expect(() => toolPairingBalancedAfter(session, missing)).toThrow(/surface seq 999 not found/)
session.append('user/message', {
content: [{ type: 'text', text: 'first node after empty cache' }],
source: { kind: 'user' },
}, SURFACE)
expect(toolPairingBalancedAfter(session, session.surface.nodes[0]!)).toBe(true)
})
})
describe('tool-pairing cache refresh', () => {
it('does no event reads for unchanged or log-only growth, folds only appended nodes, and rebuilds on replace', () => {
const events: SessionEvent[] = [
{
type: 'user/message', seq: 0, time: 0,
data: { content: [{ type: 'text', text: 'user' }], source: { kind: 'user' } },
surfaceOp: 'append',
},
{
type: 'assistant/message', seq: 1, time: 1,
data: { turn: 1, step: 1, content: [{ type: 'tool-call', id: CallId('c1'), name: 'one', arguments: '{}' }] },
surfaceOp: 'append',
},
{
type: 'tool/result', seq: 2, time: 2,
data: { turn: 1, step: 1, callId: CallId('c1'), content: [], isError: false },
surfaceOp: 'append',
},
]
const nodes: SurfaceNode[] = [
{ seq: 0, prev: null, next: 1 },
{ seq: 1, prev: 0, next: 2 },
{ seq: 2, prev: 1, next: null },
]
let generation = 0
let eventCollectionReads = 0
let eventIndexReads = 0
const trackedEvents = new Proxy(events, {
get(target, property, receiver) {
if (typeof property === 'string' && /^\d+$/.test(property)) eventIndexReads += 1
return Reflect.get(target, property, receiver) as unknown
},
})
const surface = {
get nodes() { return nodes },
get replaceGeneration() { return generation },
}
const session = {
surface,
get events() {
eventCollectionReads += 1
return trackedEvents
},
} as unknown as Session
expect(toolPairingBalancedAfter(session, nodes[2]!)).toBe(true)
expect(eventCollectionReads).toBe(1)
expect(eventIndexReads).toBe(3)
expect(toolPairingBalancedBefore(session, nodes[0]!)).toBe(true)
expect(toolPairingBalancedAfter(session, nodes[1]!)).toBe(false)
expect(eventCollectionReads).toBe(1)
expect(eventIndexReads).toBe(3)
events.push({
type: 'turn/end', seq: 3, time: 3,
data: { turn: 1, reason: { kind: 'completed' } },
})
expect(toolPairingBalancedAfter(session, nodes[2]!)).toBe(true)
expect(eventCollectionReads).toBe(1)
expect(eventIndexReads).toBe(3)
events.push({
type: 'user/message', seq: 4, time: 4,
data: { content: [{ type: 'text', text: 'tail' }], source: { kind: 'user' } },
surfaceOp: 'append',
})
nodes.push({ seq: 4, prev: 2, next: null })
expect(toolPairingBalancedAfter(session, nodes[3]!)).toBe(true)
expect(eventCollectionReads).toBe(2)
expect(eventIndexReads).toBe(4)
events.push(
{
type: 'assistant/message', seq: 5, time: 5,
data: { turn: 2, step: 1, content: [{ type: 'tool-call', id: CallId('c2'), name: 'two', arguments: '{}' }] },
surfaceOp: 'append',
},
{
type: 'tool/result', seq: 6, time: 6,
data: { turn: 2, step: 1, callId: CallId('c2'), content: [], isError: false },
surfaceOp: 'append',
},
)
nodes.push(
{ seq: 5, prev: 4, next: 6 },
{ seq: 6, prev: 5, next: null },
)
expect(toolPairingBalancedAfter(session, nodes[5]!)).toBe(true)
expect(eventCollectionReads).toBe(3)
expect(eventIndexReads).toBe(6)
events.push({
type: 'user/message', seq: 7, time: 7,
data: { content: [{ type: 'text', text: 'replacement' }], source: { kind: 'user' } },
surfaceOp: { op: 'replace', start: 0, end: 6 },
})
nodes.splice(0, nodes.length, { seq: 7, prev: null, next: null })
generation += 1
expect(toolPairingBalancedAfter(session, nodes[0]!)).toBe(true)
expect(eventCollectionReads).toBe(4)
expect(eventIndexReads).toBe(7)
})
it('rebuilds defensively when a same-generation surface node count regresses', () => {
const events: SessionEvent[] = [
{
type: 'user/message', seq: 0, time: 0,
data: { content: [], source: { kind: 'user' } }, surfaceOp: 'append',
},
{
type: 'user/message', seq: 1, time: 1,
data: { content: [], source: { kind: 'user' } }, surfaceOp: 'append',
},
]
const nodes: SurfaceNode[] = [
{ seq: 0, prev: null, next: 1 },
{ seq: 1, prev: 0, next: null },
]
const session = {
events,
surface: { nodes, replaceGeneration: 0 },
} as unknown as Session
expect(toolPairingBalancedAfter(session, nodes[1]!)).toBe(true)
nodes.pop()
expect(toolPairingBalancedAfter(session, nodes[0]!)).toBe(true)
})
})
describe('tool-pairing corrupt surfaces', () => {
it('throws for an orphan result during a rebuild', () => {
const session = new Session(SessionId('orphan-rebuild'))
session.append('tool/result', {
turn: 1, step: 1, callId: CallId('orphan'), content: [], isError: false,
}, SURFACE)
expect(() => toolPairingBalancedAfter(session, session.surface.nodes[0]!)).toThrow(/no matching tool-call/)
})
it('retries an orphan result in an appended tail without committing partial cache state', () => {
const session = new Session(SessionId('orphan-tail'))
session.append('user/message', {
content: [{ type: 'text', text: 'safe head' }], source: { kind: 'user' },
}, SURFACE)
expect(toolPairingBalancedAfter(session, session.surface.nodes[0]!)).toBe(true)
session.append('tool/result', {
turn: 1, step: 1, callId: CallId('orphan'), content: [], isError: false,
}, SURFACE)
expect(() => toolPairingBalancedAfter(session, session.surface.nodes[1]!)).toThrow(/no matching tool-call/)
expect(() => toolPairingBalancedAfter(session, session.surface.nodes[1]!)).toThrow(/no matching tool-call/)
})
it('throws when a current surface seq has no matching event or indexes the wrong event', () => {
const missingNode: SurfaceNode = { seq: 1, prev: null, next: null }
const missing = {
events: [{
type: 'user/message', seq: 0, time: 0,
data: { content: [], source: { kind: 'user' } }, surfaceOp: 'append',
} satisfies SessionEvent],
surface: { nodes: [missingNode], replaceGeneration: 0 },
} as unknown as Session
expect(() => toolPairingBalancedBefore(missing, missingNode)).toThrow(/no matching session event/)
const mismatchedNode: SurfaceNode = { seq: 0, prev: null, next: null }
const mismatched = {
events: [{
type: 'user/message', seq: 99, time: 0,
data: { content: [], source: { kind: 'user' } }, surfaceOp: 'append',
} satisfies SessionEvent],
surface: { nodes: [mismatchedNode], replaceGeneration: 0 },
} as unknown as Session
expect(() => toolPairingBalancedBefore(mismatched, mismatchedNode)).toThrow(/no matching session event/)
})
})

View File

@@ -79,7 +79,7 @@ Every `SessionEvent` carries two optional top-level fields (structural metadata)
- Persistence plugins: subscribe to `session/event` (write-behind) and drain on `session/flush` (awaited) and fiber dispose. A durable backend reads the log and reloads it into a live session; the metadata seam (`SessionHeader`, `session.header`) is what such a backend stores beside the log.
- Replay/fork: `create(id, { seed })` validates and freezes a contiguous log and rebuilds its surface. `fork(source, boundary?, childSessionId?)` selects a completed-turn prefix and records lineage.
- Compaction: the `dsh-compact-basic` plugin appends a `user/message` with `surfaceOp: { op: 'replace', start, end }` to shadow old surface nodes behind a summary checkpoint.
- Compaction: the `dsh-compact-basic` plugin appends a `user/message` with `surfaceOp: { op: 'replace', start, end }` to shadow old surface nodes behind a summary checkpoint. Tool-pairing boundary policy and its cache belong to the [`dsh-compact` seam](../../compact/compact/README.md), while this package owns surface membership, positional links, and `replaceGeneration`.
## Model Experience

View File

@@ -24,7 +24,6 @@ export type { JsonValue } from './json.ts'
export { interruptedTurnClosers } from './repair.ts'
export type { SurfaceFoldReplacement, SurfaceFoldResult, SurfaceNode } from './surface.ts'
export { foldSurface, isSurfaceEvent, isSurfaceEligibleType } from './surface.ts'
export { isToolPairingBalanced } from './tool-pairing.ts'
export { applyHeaderDelta, canonicalHeader, diffHeader, foldRequestHeader, headerEquals } from './request-header.ts'
declare module 'cordis' {

View File

@@ -1,56 +0,0 @@
/**
* Tool-pairing balance over a session surface. Compaction changes surface
* positions, so safe cuts are derived from tool-call/result content on the
* surface rather than step markers in the append-only log.
* @module @deepseek-ai/dsh-session/tool-pairing
*/
import type { SessionEvent } from './types.ts'
import type { SurfaceNode } from './surface.ts'
/**
* The tool-pairing delta of a surface node: how it shifts the count of
* unanswered tool calls. An `assistant/message` opens one bracket per
* `tool-call` block; a `tool/result` closes one; every other surface node
* (`user/message`, `context/message`, `steering/message`, a usage-only
* `assistant/message` with no tool-call blocks) is pairing-neutral.
*/
function nodeDelta(event: SessionEvent): number {
switch (event.type) {
case 'assistant/message':
return event.data.content.filter(block => block.type === 'tool-call').length
case 'tool/result':
return -1
// Non-pairing surface nodes and every non-surface event contribute nothing.
default:
return 0
}
}
/**
* Check that a surface cut does not split a tool call from its result. A region
* is safe to collapse only when the cuts before its first node and after its
* last node both return `true`.
* @param nodes - the surface linked list in head→tail order.
* @param events - the session log each node's `seq` indexes into.
* @param beforeSeq - node immediately after the cut; `null` or a seq absent from the surface means after-tail.
* @returns whether every call before the cut has its result before the cut.
* @throws if a result appears without a preceding open call.
*/
export function isToolPairingBalanced(
nodes: readonly SurfaceNode[],
events: readonly SessionEvent[],
beforeSeq: number | null,
): boolean {
let depth = 0
for (const node of nodes) {
if (node.seq === beforeSeq) return depth === 0
// eslint-disable-next-line @typescript-eslint/no-non-null-assertion
depth += nodeDelta(events[node.seq]!)
if (depth < 0) {
throw new Error(`tool-pairing balance: tool/result at surface seq ${node.seq} has no matching tool-call (corrupt surface)`)
}
}
// A missing cut node means the after-tail boundary.
return depth === 0
}

View File

@@ -1,293 +0,0 @@
import { describe, expect, it } from 'vitest'
import { CallId } from '@deepseek-ai/dsh-llm'
import { Session, SessionId, isToolPairingBalanced } from '../src/index.ts'
import type { SessionEvent, SurfaceNode } from '../src/index.ts'
/**
* Unit coverage for compaction-cut safety: a cut is balanced only when it
* separates no assistant tool call from its result. Non-step nodes are neutral,
* and replace operations prove surface order—not raw log order—is authoritative.
*/
const SURFACE = { surfaceOp: 'append' as const }
/** Surface nodes + log for a session, the two args the balance check takes. */
function surfaceOf(session: Session): { nodes: readonly SurfaceNode[]; events: readonly SessionEvent[] } {
return { nodes: session.surface.nodes, events: session.events }
}
/** The cut BEFORE the surface node at `seq` is balanced (safe region start). */
function startBalanced(session: Session, seq: number): boolean {
const { nodes, events } = surfaceOf(session)
return isToolPairingBalanced(nodes, events, seq)
}
/** The cut AFTER the surface node at `seq` is balanced (safe region end). */
function endBalanced(session: Session, seq: number): boolean {
const { nodes, events } = surfaceOf(session)
const node = nodes.find(n => n.seq === seq)
if (!node) throw new Error(`seq ${seq} is not a surface node`)
return isToolPairingBalanced(nodes, events, node.next)
}
/** Surface seq of the nth (0-based) event of a given type. */
function seqOf(s: Session, type: SessionEvent['type'], nth = 0): number {
return s.events.filter(e => e.type === type)[nth]!.seq
}
/** A closed turn with one closed step holding an assistant + its tool result. */
function toolStepSession(): Session {
const s = new Session(SessionId('tool-step'))
s.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
s.append('user/message', { content: [{ type: 'text', text: 'go' }], source: { kind: 'user' } }, SURFACE)
s.append('step/start', { turn: 1, step: 1 })
s.append('assistant/message', {
turn: 1, step: 1,
content: [
{ type: 'text', text: 'calling' },
{ type: 'tool-call', id: CallId('c1'), name: 'bash', arguments: '{}' },
],
}, SURFACE)
s.append('tool/call', { turn: 1, step: 1, callId: CallId('c1'), name: 'bash', arguments: '{}' })
s.append('tool/result', { turn: 1, step: 1, callId: CallId('c1'), content: [{ type: 'text', text: 'out' }], isError: false }, SURFACE)
s.append('step/end', { turn: 1, step: 1 })
s.append('turn/end', { turn: 1, reason: { kind: 'completed' } })
return s
}
describe('isToolPairingBalanced — region START (cut before a node)', () => {
it('is true for a pre-step user/message (belongs to no step)', () => {
const s = toolStepSession()
expect(startBalanced(s, seqOf(s, 'user/message'))).toBe(true)
})
it('is true for the first surface node of a step (the assistant/message)', () => {
// The cut before the assistant is balanced — nothing unanswered precedes it.
const s = toolStepSession()
expect(startBalanced(s, seqOf(s, 'assistant/message'))).toBe(true)
})
it('is false for a tool/result whose assistant/message precedes it in the same step', () => {
// The cut before the tool/result has one unanswered tool-call (the
// assistant's) → starting the region here would orphan that call.
const s = toolStepSession()
expect(startBalanced(s, seqOf(s, 'tool/result'))).toBe(false)
})
it('is true at the surface head (nothing precedes)', () => {
const s = new Session(SessionId('lone'))
s.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
s.append('user/message', { content: [{ type: 'text', text: 'hi' }], source: { kind: 'user' } }, SURFACE)
expect(startBalanced(s, seqOf(s, 'user/message'))).toBe(true)
})
})
describe('isToolPairingBalanced — region END (cut after a node)', () => {
it('is true for the last surface node of a closed step (the tool/result)', () => {
// After the tool/result the assistant's single call is answered → balanced.
const s = toolStepSession()
expect(endBalanced(s, seqOf(s, 'tool/result'))).toBe(true)
})
it('is false for an assistant/message with a later tool/result in the same step', () => {
// After the assistant its tool-call is still unanswered → ending here strands
// the result.
const s = toolStepSession()
expect(endBalanced(s, seqOf(s, 'assistant/message'))).toBe(false)
})
it('is true for a pre-step user/message', () => {
const s = toolStepSession()
expect(endBalanced(s, seqOf(s, 'user/message'))).toBe(true)
})
it('is false at the tail when the node is inside an open (unclosed) step', () => {
// step/start then an assistant tool-call, but no tool/result yet (mid-flight).
// The after-tail cut still has one unanswered call → not balanced.
const s = new Session(SessionId('open-step'))
s.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
s.append('step/start', { turn: 1, step: 1 })
s.append('assistant/message', {
turn: 1, step: 1,
content: [{ type: 'tool-call', id: CallId('c1'), name: 'bash', arguments: '{}' }],
}, SURFACE)
expect(endBalanced(s, seqOf(s, 'assistant/message'))).toBe(false)
})
it('is true at the tail when the node is a trailing inter-step node (step already closed)', () => {
// A steering message appended after step/end, at the tail. The prior step's
// pair is balanced and steering is neutral → the after-tail cut is balanced.
const s = new Session(SessionId('trailing-steer'))
s.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
s.append('step/start', { turn: 1, step: 1 })
s.append('assistant/message', { turn: 1, step: 1, content: [{ type: 'text', text: 'a' }] }, SURFACE)
s.append('step/end', { turn: 1, step: 1 })
s.append('steering/message', { turn: 1, content: [{ type: 'text', text: 's' }], source: { kind: 'user' } }, SURFACE)
expect(endBalanced(s, seqOf(s, 'steering/message'))).toBe(true)
})
it('is true at the tail when no step ever opened', () => {
const s = new Session(SessionId('no-step'))
s.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
s.append('user/message', { content: [{ type: 'text', text: 'hi' }], source: { kind: 'user' } }, SURFACE)
expect(endBalanced(s, seqOf(s, 'user/message'))).toBe(true)
})
})
describe('isToolPairingBalanced — multiple tool calls in one assistant message', () => {
// An assistant message with two tool-calls needs BOTH results before the cut
// after it is balanced — depth +2, then -1, -1.
function twoCallStep(): Session {
const s = new Session(SessionId('two-call'))
s.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
s.append('step/start', { turn: 1, step: 1 })
s.append('assistant/message', {
turn: 1, step: 1,
content: [
{ type: 'tool-call', id: CallId('c1'), name: 'a', arguments: '{}' },
{ type: 'tool-call', id: CallId('c2'), name: 'b', arguments: '{}' },
],
}, SURFACE)
s.append('tool/result', { turn: 1, step: 1, callId: CallId('c1'), content: [{ type: 'text', text: '1' }], isError: false }, SURFACE)
s.append('tool/result', { turn: 1, step: 1, callId: CallId('c2'), content: [{ type: 'text', text: '2' }], isError: false }, SURFACE)
s.append('step/end', { turn: 1, step: 1 })
s.append('turn/end', { turn: 1, reason: { kind: 'completed' } })
return s
}
it('is unbalanced after the first of two results (one call still open)', () => {
const s = twoCallStep()
expect(endBalanced(s, seqOf(s, 'tool/result', 0))).toBe(false)
})
it('is balanced after the second result (both calls answered)', () => {
const s = twoCallStep()
expect(endBalanced(s, seqOf(s, 'tool/result', 1))).toBe(true)
})
})
describe('isToolPairingBalanced — a mid-step injection context/message', () => {
// The injected context is pairing-neutral, but both adjacent cuts remain
// unbalanced because the tool call is still open across them.
function midStepInjection(): Session {
const s = new Session(SessionId('mid-inject'))
s.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
s.append('step/start', { turn: 1, step: 1 })
s.append('assistant/message', {
turn: 1, step: 1,
content: [{ type: 'tool-call', id: CallId('c1'), name: 'bash', arguments: '{}' }],
}, SURFACE)
s.append('context/message', { content: [{ type: 'text', text: 'bg task done' }], source: { kind: 'plugin', plugin: 'tool-bash' } }, SURFACE)
s.append('tool/result', { turn: 1, step: 1, callId: CallId('c1'), content: [{ type: 'text', text: 'out' }], isError: false }, SURFACE)
s.append('step/end', { turn: 1, step: 1 })
s.append('turn/end', { turn: 1, reason: { kind: 'completed' } })
return s
}
it('start cut before the mid-step context/message is unbalanced (call still open)', () => {
const s = midStepInjection()
expect(startBalanced(s, seqOf(s, 'context/message'))).toBe(false)
})
it('end cut after the mid-step context/message is unbalanced (call still open)', () => {
const s = midStepInjection()
expect(endBalanced(s, seqOf(s, 'context/message'))).toBe(false)
})
})
describe('isToolPairingBalanced on an injection turn (no step)', () => {
// An idle inject() wraps a context/message in a bare turn/start →
// context/message → turn/end with NO step. The context node is a free boundary
// both ways (pairing-neutral, nothing open around it).
function injectionSession(): Session {
const s = new Session(SessionId('injection'))
s.append('turn/start', { turn: 1, trigger: { kind: 'injection', source: { kind: 'user' } } })
s.append('context/message', { content: [{ type: 'text', text: 'ctx' }], source: { kind: 'user' } }, SURFACE)
s.append('turn/end', { turn: 1, reason: { kind: 'completed' } })
return s
}
it('start: balanced', () => {
const s = injectionSession()
expect(startBalanced(s, seqOf(s, 'context/message'))).toBe(true)
})
it('end: balanced', () => {
const s = injectionSession()
expect(endBalanced(s, seqOf(s, 'context/message'))).toBe(true)
})
})
describe('isToolPairingBalanced — CBR-001: a head checkpoint left by a replace op', () => {
// A replacement checkpoint has a high log seq but sits at the surface head;
// its cuts are balanced regardless of later raw-log neighbors.
function checkpointHeadedSession(): Session {
const s = new Session(SessionId('checkpoint'))
// A closed turn with a tool step → surface [u1, asst(call), result].
s.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
s.append('step/start', { turn: 1, step: 1 })
s.append('user/message', { content: [{ type: 'text', text: 'u1' }], source: { kind: 'user' } }, SURFACE)
s.append('assistant/message', {
turn: 1, step: 1,
content: [{ type: 'tool-call', id: CallId('c1'), name: 'bash', arguments: '{}' }],
}, SURFACE)
s.append('tool/result', { turn: 1, step: 1, callId: CallId('c1'), content: [{ type: 'text', text: 'out' }], isError: false }, SURFACE)
s.append('step/end', { turn: 1, step: 1 })
s.append('turn/end', { turn: 1, reason: { kind: 'completed' } })
// An OPEN turn whose step is in progress (loop fires compaction here).
s.append('turn/start', { turn: 2, trigger: { kind: 'message', source: { kind: 'user' } } })
s.append('step/start', { turn: 2, step: 1 })
// Compaction replaces the whole turn-1 surface ([u1, asst, result]) with one
// summary user/message — appended now, so it carries a high log seq.
const u1 = seqOf(s, 'user/message')
const result = s.events.find(e => e.type === 'tool/result')!.seq
const shadowedSeqs = s.surface.nodes.map(node => node.seq)
s.append('user/message', {
content: [{ type: 'text', text: 'CHECKPOINT' }],
source: { kind: 'plugin', plugin: 'compact' },
}, { surfaceOp: { op: 'replace', start: u1, end: result }, sourceEventSeqs: shadowedSeqs })
// The step's own assistant/message lands AFTER the checkpoint in the log,
// still inside the open step.
s.append('assistant/message', { turn: 2, step: 1, content: [{ type: 'text', text: 'a2' }] }, SURFACE)
return s
}
it('the head checkpoint sits at the surface head while a later surface node follows it in the log', () => {
const s = checkpointHeadedSession()
const nodes = s.surface.nodes
const checkpointSeq = nodes[0]!.seq
// The checkpoint heads the surface, yet a surface node (the open step's
// assistant) follows it in LOG order — the exact split between surface
// position and log position that the log-position scan tripped on.
const laterSurfaceInLog = s.events.find(
e => e.seq > checkpointSeq && nodes.some(n => n.seq === e.seq),
)
expect(laterSurfaceInLog).toBeDefined()
expect(nodes[0]!.seq).toBe(checkpointSeq)
})
it('start cut before the head checkpoint is balanced (it is the head)', () => {
const s = checkpointHeadedSession()
expect(startBalanced(s, s.surface.nodes[0]!.seq)).toBe(true)
})
it('end cut after the head checkpoint is balanced (it carries no tool pair)', () => {
// This is the exact assertion the log-position scan failed: the forward log scan from the
// checkpoint reached the open step's assistant/message and wrongly reported mid-step.
const s = checkpointHeadedSession()
expect(endBalanced(s, s.surface.nodes[0]!.seq)).toBe(true)
})
})
describe('isToolPairingBalanced — corrupt surface guard', () => {
it('throws when a tool/result has no preceding tool-call (depth goes negative)', () => {
// A surface that opens with a tool/result (no assistant call before it) is
// structurally corrupt — surfaced loudly rather than mis-classified.
const s = new Session(SessionId('corrupt'))
s.append('turn/start', { turn: 1, trigger: { kind: 'message', source: { kind: 'user' } } })
s.append('step/start', { turn: 1, step: 1 })
s.append('tool/result', { turn: 1, step: 1, callId: CallId('c1'), content: [{ type: 'text', text: 'x' }], isError: false }, SURFACE)
const { nodes, events } = surfaceOf(s)
expect(() => isToolPairingBalanced(nodes, events, null)).toThrow(/no matching tool-call/)
})
})

View File

@@ -6,7 +6,7 @@
Abstract compaction service. Implementations own token estimation, retention, and summarization, but a successful run must replace the selected surface span with one summary node and prevent concurrent compaction of the same session. Load one implementation per context as `ctx.compact`.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/compact/compact/src/index.ts#L36)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/compact/compact/src/index.ts#L37)
### ctx.compact.compactIfNeeded(agent, fullSystemPrompt, sessionPrefix, signal)
@@ -23,7 +23,7 @@ Check token pressure and compact if the conversation is too large. Estimate the
**Returns** the compaction result, or `null` if no compaction was needed.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/compact/compact/src/index.ts#L56)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/compact/compact/src/index.ts#L57)
### ctx.compact.compactRegion(session, start, end, agent, signal?)
@@ -31,7 +31,7 @@ Check token pressure and compact if the conversation is too large. Estimate the
abstract compactRegion( session: Session, start: number, end: number, agent: CompactAgentContext, signal?: AbortSignal, ): Promise<CompactionResult>
```
Forcibly compact a range of surface nodes into a single summary node. `start` and `end` name an inclusive span by surface position, not numeric seq order; replacements can make visible seqs non-monotonic. Both edges must be balanced so assistant tool calls remain paired with their results. A model- backed implementation forwards cancellation and rejects active, missing, reversed, or unbalanced ranges.
Forcibly compact a range of surface nodes into a single summary node. `start` and `end` name an inclusive span by surface position, not numeric seq order; replacements can make visible seqs non-monotonic. Both edges must be balanced so assistant tool calls remain paired with their results. A model- backed implementation forwards cancellation and rejects active, missing, reversed, or unbalanced ranges. Use toolPairingBalancedBefore and toolPairingBalancedAfter for the edge checks.
- `session` — session to mutate.
- `start` — first surface seq, inclusive.
@@ -41,4 +41,4 @@ Forcibly compact a range of surface nodes into a single summary node. `start` an
**Returns** the replaced range and summary.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/compact/compact/src/index.ts#L79)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/compact/compact/src/index.ts#L82)

View File

@@ -307,7 +307,7 @@ Creation announcement during session publication. A synchronous throw vetoes and
- `session` — the session just entered and announced.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L47)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L46)
### session/disposed
@@ -321,7 +321,7 @@ Emitted once when an announced session leaves the store, including publication r
- `session` — the session that is no longer live in the store.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L57)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L56)
### session/event
@@ -336,7 +336,7 @@ Post-commit, fire-and-forget append feed. The listener snapshot resolves before
- `session` — the session whose log grew.
- `event` — the appended event, exactly as recorded.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L69)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L68)
### session/flush
@@ -350,7 +350,7 @@ Awaited parallel durability checkpoint: every listener runs and the caller await
- `session` — the session whose buffered events must reach durable storage.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L79)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L78)
## subagent/*

View File

@@ -7,7 +7,7 @@
In-memory session store (`ctx.sessions`).
Persistence is intentionally not implemented here — persistence plugins subscribe to `session/event` and flush on `session/flush` / dispose.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L540)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L539)
### ctx.sessions.create(id?, options?)
@@ -23,7 +23,7 @@ For an agent whose session must be torn down IN ORDER with its loop (so the loop
**Returns** the live session, already entered and announced.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L569)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L568)
### ctx.sessions.prepare(id?, options?)
@@ -38,7 +38,7 @@ Build a session WITHOUT entering it into the store — validate the id/cwd and c
**Returns** the constructed session, NOT yet in the store.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L598)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L597)
### ctx.sessions.enter(session)
@@ -53,7 +53,7 @@ Re-checks the id for a duplicate: `prepare` and `enter` are public cross-package
**Returns** the detach disposer (publication hooks + store removal). When called from a synchronous `session/created` listener, removal and disposal wait until that creation dispatch unwinds.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L642)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L641)
### ctx.sessions.announce(session)
@@ -65,7 +65,7 @@ Emit `session/created` exactly once for an entered session (with the carrier ent
- `session` — the entered session to announce to listeners.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L697)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L696)
### ctx.sessions.flush(session)
@@ -79,7 +79,7 @@ Dispatch the awaited `session/flush` durability checkpoint for `session`, with t
**Returns** resolves when every flush listener has settled; after all settle, rejects with the first registered listener failure if any listener failed.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L749)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L748)
### ctx.sessions.get(id)
@@ -93,7 +93,7 @@ Look up a live session.
**Returns** the session, or undefined when no live session has that id.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L781)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L780)
### ctx.sessions.list()
@@ -105,7 +105,7 @@ All live sessions, in creation order.
**Returns** a fresh array; mutating it does not affect the store.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L789)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L788)
### ctx.sessions.fork(source, boundary?, childSessionId?)
@@ -121,4 +121,4 @@ Create a live child session from a turn-enclosed prefix of a live source. `bound
**Returns** The created live child session.
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L806)
[Source](https://github.com/deepseek-harness/deepseek-harness/blob/master/packages/core/session/src/index.ts#L805)