dsh-lcx-codex 0.4.2 → 0.4.3-pre.13

package/CHANGELOG.md

@@ -1,224 +0,0 @@
-# Changelog
-
-## 0.4.2-pre.1 - candidate under runtime acceptance
-
-### Architecture
-
-- Make the LCX master switch the GPT Responses lifecycle ownership switch: when enabled, the first ordinary Agent turn, tools, Native V2 compaction, native replay and portable GPT-model migration share one LCX final-wire path.
-- Upgrade only the plugin's direct Pi dependency to `0.84.3`; DSH 0.1.1-rc.2 and its host Pi 0.82.x remain isolated and unchanged.
-- Reuse Pi 0.84 canonical Responses serializers and stream semantics for strict/grammar/custom tools, deferred `additional_tools` / `tool_search`, namespace, reasoning, IDs and cache behavior.
-- Add one standard Responses request builder and shared LCX HTTP/SSE transport/parser bridge; remove the old portable recursive bypass back to the DSH adapter.
-- Preserve checkpoint v5, DSH compaction transactions, Native-first pressure coordination, first-checkpoint Basic fallback, restart/resume and Search/#20 architecture.
-- Normalize ordinary/replay failures into DSH terminal taxonomy and keep DSH as the visible retry owner.
-
-### Validation state
-
-- Engineering gates pass on the private work branch; exact commit-bound DSH + Sub2API runtime acceptance remains mandatory before integration, publication or a `VERIFIED` compatibility claim.
-
-## 0.4.1 - 2026-08-25
-
-### Stable promotion
-
-- Promote the fully QA-passed `0.4.1-pre.1` runtime to stable `0.4.1` with **zero functional/runtime code changes**.
-- Make Alpha stateful continuation/fail-closed fixes, DSH compatibility-seam isolation, typed protocol hardening, conservative token budgeting, and the installed Settings lifecycle fix part of the default stable line.
-- Publish stable `0.4.1` to npm dist-tag `latest`; the prior `prelatest=0.4.1-pre.1` remains historical prerelease state.
-
-### Validation basis
-
-- `0.4.1-pre.1` passed Installed Candidate + Cross-feature QA before publication.
-- Public promotion preserved runtime/source parity with the frozen prerelease candidate.
-- Stable promotion requires package/version, typecheck, tests, schema, package/import boundary and zero-runtime-diff checks before publication.
-
-## 0.4.1-pre.1 - 2026-08-25
-
-### Released prerelease
-
-- Published as `dsh-lcx-codex@0.4.1-pre.1` on npm dist-tag `prelatest`; stable `latest` remained `0.4.0` at prerelease publication time.
-- Integrates the accepted Alpha stateful continuation and fail-closed capability work from #20, #21, and #22.
-- Includes accepted compatibility seam isolation (#24), typed protocol-core hardening (#25), unified conservative token budgeting (#26), and the installed-plugin settings lifecycle fix (#36).
-- Passed installed-candidate and cross-feature QA before publication, including ordinary/Advanced Hosted Search, Native compaction/replay, restart/resume, model continuation, fork isolation, and fail-closed Alpha behavior where unsupported.
-- Public promotion preserved source/runtime parity with the frozen private candidate; package, schema, typecheck, import, install-chain, and publication gates passed.
-- Post-release registry installation and Loader/client/settings-chain verification passed. The final provider-gated ordinary `web_search` recheck was not covered because the temporary authorized upstream credential returned HTTP 401; this was recorded as an external-auth verification gap rather than a package/source/runtime defect.
-- Superseded as the recommended release by stable `0.4.1` after zero-functional-change promotion.
-
-## 0.4.0 - 2026-08-23
-
-### Stable promotion
-
-- Promote the fully validated `0.4.0-rc.13` runtime to the stable `0.4.0` line with no functional/runtime code changes.
-- Make the rc.13 Native V2 hardening, canonical Responses replay/serialization fixes, session isolation, cache continuity, and concurrency/lifecycle fixes the recommended stable release.
-- Preserve `0.3.4` only as historical stable state; `0.4.0` becomes the default npm `latest` after publication.
-
-### Validation
-
-- Runtime-sensitive source is required to remain content-identical to the already live-tested rc.13 release.
-- Full suite: 58/58 passed; DSH schema validation: 4/4 passed.
-- Real rc.13 DSH acceptance already covers automatic Native pressure Compact, replay/cache re-warm, restart/resume, real GUI `/compact`, parent/child opaque isolation, and dual-session same-preset ServiceMutex failure cleanup.
-- Stable promotion adds no new runtime behavior; publication is a version/documentation promotion only.
-
-## 0.4.0-rc.13 - 2026-08-23
-
-### Fixed
-
-- Restrict opaque Native checkpoint replay to the exact source DSH session. Parent/child ancestry remains valid only for portable migration, so a fork never sends the parent session's opaque Native state.
-- Reconcile Responses replay deltas and terminal output by stable item/call identity instead of `output_index` alone, preventing changed-index duplicate/empty text blocks and keeping terminal-only function calls balanced.
-- Replace the simplified Native message/tool serializer with the public `@earendil-works/pi-ai@0.82.1` OpenAI Responses converters, preserving reasoning/message identities, tool linkage, constrained-tool semantics, deferred-tool transcript semantics, and DSH image boundaries.
-- Match ordinary Pi `openai-responses` system-prompt placement: Native compact/replay now carries the developer/system prelude inside the canonical input prefix rather than introducing a separate top-level `instructions` prefix difference.
-- Restore strict validation of DSH Pi replay-state envelopes before reusing native signatures; mismatched replay metadata degrades to a portable foreign-assistant projection instead of injecting stale provider-native identity.
-- Treat Pi canonical role-only developer/user items as durable retained history and keep image persistence on `dsh_image_attachment` references rather than request image payloads.
-- Use Pi/model or explicitly configured DSH Responses compatibility only; unknown custom routes no longer assume strict tools, grammar tools, or tool-search support.
-- Mirror DSH/OpenAI Responses generation controls into Native compact/replay (`reasoning` + encrypted-reasoning include, temperature, and max output tokens), preventing the real Terra `xhigh` compact path from dropping the envelope used by ordinary Pi requests.
-- Source automatic-compaction generation controls from the matching session request header rather than the Basic compaction summarizer request, so the selected conversation effort (for example Terra `xhigh`) reaches the Native provider request.
-- Declare the DSH `sessions` service as an explicit Cordis runtime injection because checkpoint/replay and automatic-compaction generation parity read the live session/request header.
-- Maintain a deterministic per-session request-header cache from DSH `session/event`; `compaction/start` synchronously refreshes it from the live Session before the Native request, avoiding reliance on async-context propagation for generation parity.
-- Register that cross-session `session/event` observer with Cordis `{ global: true }`, matching DSH system-wide observers so Agent-carrier session events reach the standing plugin scope.
-- Snapshot the exact `agentArg.session.requestHeader()` inside the Native-first pressure wrapper immediately before delegating to Basic compaction; this is the deterministic pressure-path source of generation controls, with the global session-event cache retained only as a secondary path.
-- Seed the request-header cache from already-live sessions at plugin installation, seed newly announced resume/fork sessions, and evict disposed session entries so manual/non-pressure compaction remains restart-safe without retaining stale headers.
-- Match the explicit Remote V2 tool-control contract used by current Codex and mature Responses compaction implementations: Native compact/replay send `tool_choice: "auto"` and `parallel_tool_calls: true` on the plain `openai-responses` route.
-- Preserve only bounded provider machine diagnostics (`code`, `type`, `param`) from `response.failed`; the safe identifiers are included in the generic failure message/log so DSH compaction history can diagnose failures, while provider messages/bodies remain excluded.
-- Persist Pi replay envelope v2 on successful Native replay finish so DSH retains response id/stop reason plus text/reasoning native metadata across post-compact turns.
-- Preserve normal Pi OpenAI function-call identity as `call_id|item_id` for replay deltas and completed tool-call blocks, preventing loss of the provider `fc_*` item id on the next canonical request.
-- Treat the session header `config` as the effective ordinary-request envelope even when DSH marks a value in `adapterDefaults`; those materialized defaults (for example Terra `xhigh` and `maxTokens`) are still sent by normal Pi and therefore must be mirrored by Native compact/replay.
-
-### Validation
-
-- `tests/rc13-regressions.test.mjs`: 15/15 passed.
-- Full local suite: 58/58 passed; DSH schema validation: 4/4 passed; `git diff --check` passed.
-- Installed DSH/NewAPI acceptance passed on Terra xhigh: automatic 90% Native V2 compact, continuous replay/cache re-warm, DSH Web restart/resume, real GUI `/compact` plus continuation, and parent→child fork portable isolation all completed without pre-Native stock prune or cross-session opaque replay.
-- This remains a local release candidate only; no npm/GitHub publication or tag is implied.
-
-## 0.4.0-rc.12 - 2026-08-22
-
-### Fixed
-
-- Serialize every `compactIfNeeded()` call per concrete preset compaction service so same-generation sessions cannot observe each other’s temporary Native-first pruner/config state; queued calls are abortable and plugin cleanup drains active owners before restoring the original method.
-- Align Native `openai-responses` session affinity with the active Pi adapter: default OpenAI-format routes use `session_id` plus `x-client-request-id`, while OpenRouter-format routes use `x-session-id`; explicit affinity headers remain authoritative.
-- Restrict Basic fallback to allowlisted retryable first-checkpoint failures and fail closed once a Native/legacy checkpoint already exists.
-- Require a real `response.completed` terminal event with `status=completed` for Native compaction/replay and reject orphan `function_call_output` items.
-- Set credential-bearing fetches to `redirect: error` and keep provider response bodies/messages out of surfaced/logged transport errors.
-
-## 0.4.0-rc.11 - 2026-08-22
-
-### Fixed
-
-- Restore ERR-051 cache/session affinity semantics that regressed in the rc.8 refactor: Native V2 compaction and same-route replay now use the DSH/Pi conversation session id as the clamped `prompt_cache_key` instead of a route fingerprint, inherit the active provider `cacheRetention`, emit `prompt_cache_retention: 24h` only for supported `long` retention, and omit Native cache affinity when retention is `none`.
-- Keep ordinary Hosted Search on its intentionally separate `dsh-lcx-search:<route hash>` namespace.
-
-## 0.4.0-rc.10 - 2026-08-22
-
-### Fixed
-
-- Fix Native-first pressure coordination for DSH 0.1.1-rc.2 Agent presets by using the public `agentPresets.serviceFor(agent, name)` resolver for preset-local `compaction` and `toolResultPruner` instances. This replaces the rc.9 assumption that ordinary `agent.ctx` lookup could see entry-local isolated services.
-
-## 0.4.0-rc.9 - 2026-08-22
-
-### Fixed
-
-- Restore the Alpha capability/ref-store interfaces consumed by `lib/index.js`; the rc.8 release package could pass its unit tests but fail immediately at module import with missing `AlphaCapabilityStore` / `AlphaRefStore` exports.
-- Add package-entry import regression coverage so CI fails when the server entry point and internal module exports drift out of sync.
-- Restore Native-first pressure coordination for isolated agent presets using DSH 0.1.1-rc.2's public `agentPresets.serviceFor(agent, name)` resolver. Preset-local `compaction` / `toolResultPruner` services are not visible through ordinary host or `agent.ctx` lookup; the plugin now addresses the actual per-Agent instances before applying the 90% Native / 95% emergency policy. A root service-lifecycle hook remains for non-preset deployments.
-
-## 0.4.0-rc.8
-
-- Rebase the plugin on DSH `0.1.1-rc.2`; older DSH releases are no longer a supported runtime target.
-- Native V2 compaction image replay now uses DSH `attachments.readImageRequest()` with the active `llm-pi-ai` route's request-image pixel/byte policy instead of reading normalized master bytes directly.
-- Native image requests use DSH's deterministic `offloadRequestImagesWithPolicy()` projection before serialization, matching the current request-size behavior for long image-heavy sessions.
-- Keep rc.7 active-Agent Hosted Search routing, isolated search cache namespace, 240s search timeout, rc.5 conversation-fidelity checkpointing, and rc.6 90% Native-first / 95% emergency pressure policy unchanged.
-- CI installs against current declared DSH packages instead of enforcing the stale rc.8 lockfile.
-
-## 0.4.0-rc.7 - 2026-08-22
-
-### Fixed
-
-- Ordinary DSH `web_search` now follows the active Agent `provider/model` instead of always using the plugin fallback GPT model. A Luna conversation now searches with Luna; a Sol conversation now searches with Sol.
-- Added an AsyncLocalStorage route bridge at the DSH `tools/execute` boundary so the provider-only `ctx.web.search()` seam can receive Agent route context without changing the model-visible `web_search` schema.
-- Hosted Search now uses a separate stable `dsh-lcx-search:<route hash>` `prompt_cache_key`, avoiding intentional cache-key sharing with Native conversation replay.
-- The settings UI now labels the configured Responses endpoint/model as **fallback** values, matching their actual rc.7 role.
-
-### Kept from rc.6
-
-- 240-second default DSH `web_search` deadline.
-- Native-first automatic pressure policy: 90% Native V2, 95% emergency DSH prune.
-- rc.5 conversation-fidelity checkpoints and restart-safe DSH session-log persistence.
-
-### Docs / release
-
-- Reworked the README around the current architecture and real cache observations.
-- Added a blue/white DSH-LCX-CODEX hero banner for GitHub/npm.
-- GitHub trusted publishing is wired through `.github/workflows/publish.yml`: pre-release tags publish to npm dist-tag `next`; stable tags publish to `latest`.
-
-## 0.4.0-rc.6
-
-- Added Native-first automatic pressure coordination for GPT Responses sessions: below the configured Native threshold the plugin suppresses DSH's stock 80% pressure compaction/prune path; at the default 90% threshold it lets compaction proceed while temporarily suppressing tool-result pruning so Native V2 runs first.
-- Added a separate emergency prune threshold (default 95%). At or above this zone, DSH's replay-safe tool-result pruner is allowed to run before compaction as overflow protection.
-- Added adjustable `web_search` tool deadline, default 240 seconds (30–600s). This mutates only DSH's non-model-visible `ToolDefinition.timeoutMs`, so the model tool schema and prompt-cache prefix do not change.
-- Added Settings UI controls for automatic compaction, Native threshold, emergency prune threshold, and web search timeout.
-- Kept the rc.5 checkpoint/fidelity format unchanged (`lcx-native-compaction-v5`); rc.6 is a pressure/timeout coordination release, not another checkpoint migration.
-
-## 0.4.0-rc.5
-
-- Adds a bounded conversation-fidelity layer after real DSH testing showed assistant-only facts could be lost by opaque Native V2 compaction.
-- New `lcx-native-compaction-v5` checkpoints retain selected user/developer/system messages plus user-visible assistant final answers before the opaque compaction item.
-- Keeps explicit retained history within an estimated 64k-token ceiling; defaults reserve at most 24k for assistant answers and cap one retained assistant answer at about 3k tokens.
-- Does not copy reasoning, tool calls/results, raw search payloads, or telemetry into the fidelity prefix.
-- Repairs rc.3/rc.4 v4 checkpoints from append-only `shadowedSeqs`, including assistant-visible answers when the original DSH events still exist.
-- Stores only the single opaque compaction output item, ignoring unrelated terminal output items from nonstandard proxies.
-- Route compatibility now accepts both native checkpoint versions 4 and 5.
-- Adds regression coverage for the exact assistant-only anchor failure (`Cobalt-Sparrow-604` / `81736`) and the 64k retention ceiling.
-
-## 0.4.0-rc.4
-
-- Fixes a Native V2 replay fidelity bug found by real DSH session-log testing.
-- Native checkpoints now persist the retained client-authored Responses messages before the opaque `compaction` item, matching current OpenAI Codex remote-compaction V2 replacement-history semantics.
-- Existing rc.3 opaque-only v4 checkpoints are repaired on replay by reconstructing the missing shadowed user history from the DSH append-only session log.
-- Adds replay/retention regression coverage and an explicit Native success diagnostic.
-- Keeps the rc.3 search-provider, remote-first fallback, and session-log-native checkpoint architecture unchanged.
-
-## 0.4.0-rc.3
-
-- Fix Cordis external-package loading: all `ctx.web`/`ctx.llm` service access now occurs inside an explicit `ctx.inject(['llm', 'web'], ...)` scope.
-- This fixes `cannot get property "web" without inject` when DSH loads the plugin from a profile-installed `.tgz`.
-- No protocol or checkpoint-format changes from rc.2.
-
-## 0.4.0-rc.2 - 2026-08-21
-
-- Packaging-only fix over rc.1: remove unnecessary `@deepseek-ai/dsh-compaction-basic` and `@deepseek-ai/dsh-web` peer declarations.
-- The plugin consumes DSH runtime services through injected `ctx.*` seams and does not import or mount either package directly.
-- Avoids misleading pnpm "missing peer" warnings and, importantly, avoids encouraging users to install a second compaction backend.
-
-## 0.4.0-rc.1 - 2026-08-21
-
-### Architecture
-
-- Keep DSH `compaction-basic` as the sole compaction service owner; use only its documented/interceptable `purpose=compaction` `llm/stream` summarizer seam.
-- Replace parallel local+remote compaction with remote-first fallback.
-- Persist new Native V2 opaque state in DSH `compaction/summary.rawOutput` using `lcx-native-compaction-v4`; the model-visible replacement stays short.
-- Remove new-checkpoint writes to the v3 JSON sidecar. The v3 sidecar is now read-only compatibility for old sessions.
-- Reconstruct portable history for route migration from DSH append-only `shadowedSeqs` instead of duplicating every checkpoint's portable history.
-- Preserve same-route fork replay through DSH session ancestry.
-- Centralize the remaining direct Responses-native transport in `compact-v2.js` and `responses-replay.js`.
-
-### Search
-
-- Make `ctx.web` / DSH `web_search` the ordinary Hosted Search entry point.
-- Remove the ambiguous ordinary `websearch_gpt` tool.
-- Add opt-in `websearch_gpt_advanced` for Hosted-only controls that DSH `WebSearchRequest` cannot express.
-- Keep `websearch_alpha` independent and capability-gated.
-- Isolate the DSH rc.8 runtime SearchProvider-selection compatibility shim.
-
-### Reliability
-
-- Rehydrate DSH image attachment references when replaying legacy v3 checkpoints.
-- Resolve startup settings into runtime route config immediately instead of waiting for the first settings change.
-- Do not assume `events[seq]` is always the event whose `event.seq === seq`; use a safe fallback lookup.
-- Add protocol, session persistence, migration and architecture regression tests.
-
-### Compatibility
-
-- Node.js >= 20.
-- Target DSH `0.1.1-rc.2` only.
-- Existing 0.3.x v3 marker sessions remain best-effort readable through the old sidecar.
-
-## 0.3.1
-
-- Previous Hosted/Alpha Search and Native V2 checkpoint-v3 implementation.

package/README_EN.md

@@ -1,277 +0,0 @@
-# dsh-lcx-codex
-
-[![npm](https://img.shields.io/npm/v/dsh-lcx-codex?label=stable&color=1677ff)](https://www.npmjs.com/package/dsh-lcx-codex)
-[![CI](https://github.com/kk3ya03-star/dsh-lcx-codex/actions/workflows/publish.yml/badge.svg)](https://github.com/kk3ya03-star/dsh-lcx-codex/actions/workflows/publish.yml)
-![DSH](https://img.shields.io/badge/DSH-0.1.1--rc.2-4ea8ff)
-![Node](https://img.shields.io/badge/Node-%5E22.19%20%7C%7C%20%3E%3D24-2f855a)
-![Plugin Pi](https://img.shields.io/badge/plugin%20Pi-0.84.3-111827)
-![License](https://img.shields.io/badge/license-MIT-6b7280)
-
-[简体中文](README.md) · **English** · [Architecture](ARCHITECTURE.md) · [Changelog](CHANGELOG.md) · [Releases](https://github.com/kk3ya03-star/dsh-lcx-codex/releases)
-
-> **Keep a DSH GPT Responses conversation on one final Responses wire owner from the first ordinary turn through tools, Native V2 Compact / Replay, restart recovery, and GPT model switching.**
-
-`dsh-lcx-codex` does not replace DSH. DSH still owns the Agent, Session, model selection, tool execution, attachments, and compaction policy. When **LCX is ON**, LCX owns the final request / SSE wire for the selected GPT Responses conversation.
-
-## Current stable release
-
-**`0.4.2`** is the current stable release and a **zero-functional-change stable promotion** of the fully validated `0.4.2-pre.1` runtime.
-
-```powershell
-dsh plugin --profile web add dsh-lcx-codex
-dsh web
-```
-
-- npm dist-tag: `latest`
-- DSH: `0.1.1-rc.2`
-- Plugin Pi: `0.84.3`
-- Node.js: `^22.19.0 || >=24.0.0`
-
-`0.4.2-pre.1` remains a historical prerelease; new users should install `latest`.
-
-## Product contract
-
-```text
-LCX OFF
-= use the native DSH LLM path
-
-LCX ON
-= LCX owns the final Responses request / SSE wire
-  for the selected GPT Responses conversation starting at ordinary turn 1
-```
-
-Turn LCX off before switching to Claude, Gemini, DeepSeek, or another non-GPT model. Model switching itself does not require a DSH restart.
-
-## Why LCX exists
-
-DSH already owns the Agent, Session, tools, and compaction lifecycle. LCX addresses the final provider-native GPT Responses layer: ordinary requests, tools, Native Compact, Replay, restart/resume, and GPT route migration no longer switch final wire owners inside one conversation.
-
-```text
-                        DSH
-              Agent / Session / Tools
-                         │
-                 llm / stream seam
-                         │
-              ┌──────────┴──────────┐
-              │                     │
-           LCX OFF               LCX ON
-              │                     │
-       native DSH adapter      LCX Responses Core
-                                    │
-                         ordinary turn 1
-                                    ↓
-                                 tools
-                                    ↓
-                         Native V2 Compact
-                                    ↓
-                                Replay
-                                    ↓
-                           Restart / Resume
-                                    ↓
-                         GPT Model Migration
-```
-
-Compact therefore changes history representation without also changing request ownership.
-
-## Core capabilities
-
-### 1. Full GPT Responses lifecycle ownership
-
-With LCX ON, ordinary requests, tool continuations, Native Compact, Native Replay, restart/resume, and portable GPT migration use one LCX Responses request-builder / transport owner.
-
-DSH remains the canonical session/history owner. LCX does not create a second conversation database or replace the DSH Tool Executor.
-
-### 2. GPT-5.6 Prompt Cache
-
-`0.4.2` uses the current GPT-5.6 cache-options path and maintains stable cache identity inside a session:
-
-- implicit caching by default;
-- the supported route sends `prompt_cache_options.ttl = 30m`;
-- ordinary consecutive turns and tool-heavy workloads can reuse warm prefixes;
-- Native Compact intentionally creates a new history/cache epoch, so the old uncompressed prefix is not promised to remain reusable;
-- real long-session validation repeatedly observed near-complete prefix reuse while request topology remained stable.
-
-Cache reuse depends on provider behavior, model, request prefix, tool schemas, and session state; it is not a fixed performance guarantee.
-
-One confirmed boundary: activating a skill at runtime can change the top-level tool schema and cause a one-time cache reset. The new topology warms again on the next request. The currently supported route rejects content-level `prompt_cache_breakpoint`, and DSH `0.1.1-rc.2` does not expose authoritative dynamic-tool provenance, so `0.4.2` preserves the safe one-time reset instead of guessing tool history.
-
-### 3. Native V2 Compact + Replay
-
-DSH still decides **when to compact, what range to compact, transaction boundaries, and recovery**. LCX owns only the provider-native Responses V2 wire.
-
-Default pressure coordination:
-
-```text
-0% ─────────────────── 90% ───── 95% ───── 100%
-       normal            Native      emergency   hard cap
-                         V2          DSH prune
-```
-
-- `90%`: prefer Native V2;
-- `95%`: allow emergency DSH tool-result pruning;
-- provider-confirmed overflow stays with DSH recovery;
-- manual `/compact` keeps the native DSH compaction transaction.
-
-Native checkpoint v5 persists provider-native compaction state plus the portable retained history needed for fidelity. Compatible same-session routes can use Native replay; incompatible routes never reuse unsafe opaque state.
-
-### 4. Restart / Resume + hot GPT switching
-
-A single DSH session can continue through:
-
-```text
-Terra
-  → Compact
-  → Replay
-  → Restart DSH
-  → Resume same session
-  → Switch to Sol
-  → Continue
-```
-
-Compatible route/model resumes may restore Native opaque state. An incompatible GPT model / route drops unsafe opaque state and reconstructs portable history while remaining on LCX Responses transport.
-
-### 5. Hosted Search + stateful Web Actions
-
-Ordinary web search still uses DSH's native `web_search`. LCX can map its SearchProvider to the active GPT Hosted Search route instead of exposing a second ordinary-search tool.
-
-| Need | Entry |
-|---|---|
-| Ordinary web search | DSH `web_search` |
-| Advanced Hosted Search controls | `websearch_gpt_advanced` |
-| Stateful page / PDF actions | `websearch_alpha` |
-
-Alpha supports:
-
-```text
-search → open → find / click → screenshot
-```
-
-Alpha is off by default and is registered only after the current endpoint / provider / model / schema passes a capability probe. Unknown deployments fail closed.
-
-### 6. Isolated Pi 0.84.3 upgrade
-
-The plugin uses `@earendil-works/pi-ai 0.84.3` without overriding the DSH host dependency:
-
-```text
-DSH 0.1.1-rc.2
-└─ host Pi 0.82.1      ← unchanged
-
-dsh-lcx-codex 0.4.2
-└─ plugin Pi 0.84.3    ← isolated plugin dependency
-```
-
-Pi owns canonical Responses message/tool serialization, reasoning, IDs, strict/grammar/custom tools, `additional_tools`, `tool_search`, namespace, and stream semantics. LCX does not maintain a second generic provider framework.
-
-## Configure in 30 seconds
-
-### Requirements
-
-- Node.js `^22.19.0 || >=24.0.0`
-- DSH `0.1.1-rc.2`
-- a working GPT Responses route in DSH
-- actual upstream support for the Hosted Search / Native V2 / Alpha capabilities you enable
-
-### Suggested first-run settings
-
-| Setting | Suggested | Notes |
-|---|---:|---|
-| Enable LCX | **On** | Own the current GPT Responses conversation |
-| Use GPT Hosted Search | As needed | Use GPT Hosted Search behind DSH `web_search` |
-| Advanced Hosted Search | Off | Enable only for advanced Hosted parameters |
-| Alpha Search | Off | Enable after capability validation |
-| Native-first auto compaction | On | Use Native-first pressure coordination |
-| Native threshold | `90%` | Proactive Native V2 threshold |
-| Emergency DSH prune | `95%` | Emergency prune threshold |
-| Fallback to Basic Compaction | On | Allow bounded fallback after Native failure |
-| `web_search` timeout | `240s` | Avoid premature timeout on slower Hosted Search |
-
-### Verify observable behavior
-
-| Capability | Expected behavior |
-|---|---|
-| LCX ownership | With LCX ON, the first GPT ordinary request already uses the LCX Responses path |
-| Prompt Cache | Stable warm turns may report provider `cached_tokens` |
-| Hosted Search | Ordinary entry remains DSH `web_search` |
-| Advanced Hosted | `websearch_gpt_advanced` appears when enabled |
-| Alpha | `websearch_alpha` appears only after capability validation |
-| Native V2 | Compact produces provider-native checkpoint behavior rather than relabeling Basic Compaction |
-| Replay | The same DSH session continues after Compact and remains resumable after restart |
-
-## DSH / LCX responsibility boundary
-
-| Component | Owns |
-|---|---|
-| **DSH** | Agent loop, Session/history, GenerateOptions, model/credential selection, tool execution, AttachmentStore, pressure policy, compaction transaction |
-| **DSH compatibility seam** | Projects DSH messages / GenerateOptions into Pi Context and bridges results back to DSH |
-| **Plugin Pi 0.84.3** | Canonical Responses serialization / parser semantics |
-| **LCX** | ON/OFF ownership, final Responses body + HTTP/SSE wire, ordinary/compact/replay orchestration, Native opaque state, Search capabilities |
-
-See [ARCHITECTURE.md](ARCHITECTURE.md) for checkpoint, portable replay, cache identity, RefStore, pressure coordination, and protocol details.
-
-## Compatibility
-
-| Plugin | DSH | DSH host Pi | Plugin Pi | Status |
-|---|---|---|---|---|
-| `0.4.2` | `0.1.1-rc.2` | `0.82.1` | `0.84.3` | **VERIFIED STABLE** |
-| `0.4.1` | `0.1.1-rc.2` | `0.82.1` | `0.82.1` | historical stable |
-
-DSH `0.1.2-alpha.1` and newer Pi versions are not part of the formal `0.4.2` compatibility claim. They will be validated by affected seam rather than assumed compatible automatically.
-
-## Current boundaries
-
-- The supported route uses implicit Prompt Cache. Content-level explicit breakpoints are rejected on that route and are not exposed as a product setting.
-- Dynamic skills/plugins that change the top-level tool schema can trigger a one-time prompt-cache reset; functionality remains correct and the new topology warms again.
-- `reasoning.context` / `reasoning.mode` remain host/Pi exposure boundaries; LCX does not invent a second control surface.
-- The normal operational context profile remains in the ~`262K` class; 1.05M long context is not enabled by default.
-- Credentialed `ALPHA-004` runtime coverage remains `NOT_COVERED`, while fail-closed behavior is covered by tests.
-- Programmatic Tool Calling is not currently advertised as supported.
-
-## FAQ
-
-<details>
-<summary><strong>Why do I need to turn LCX off before switching to a non-GPT model?</strong></summary>
-
-LCX ON is a GPT Responses lifecycle ownership switch, not a universal multi-model proxy. Non-GPT models continue through native DSH adapters.
-
-</details>
-
-<details>
-<summary><strong>Why is web_search still the same tool?</strong></summary>
-
-By design. LCX changes the SearchProvider behind DSH `web_search` instead of exposing two ordinary search tools to the model.
-
-</details>
-
-<details>
-<summary><strong>Why is Alpha sometimes missing?</strong></summary>
-
-That is fail-closed behavior. `websearch_alpha` is not registered until the active route/schema passes capability probing.
-
-</details>
-
-<details>
-<summary><strong>Why can a skill load cause the cache to warm again?</strong></summary>
-
-Some skills dynamically register new top-level tools. Tool schemas are part of the cacheable prompt prefix, so a topology change can establish a new cache epoch. `0.4.2` prioritizes correct tool definitions instead of guessing provenance to force reuse of an old cache.
-
-</details>
-
-## Development and verification
-
-```bash
-npm run typecheck
-npm test
-npm run test:schema
-npm pack --ignore-scripts
-```
-
-- Architecture: [ARCHITECTURE.md](ARCHITECTURE.md)
-- Changelog: [CHANGELOG.md](CHANGELOG.md)
-- GitHub Releases: [Releases](https://github.com/kk3ya03-star/dsh-lcx-codex/releases)
-- npm: [`dsh-lcx-codex`](https://www.npmjs.com/package/dsh-lcx-codex)
-
-## License
-
-MIT
-
-> `LCX` is only the project name. This is a community project and is not affiliated with OpenAI, DeepSeek, Sub2API, or NewAPI.

package/lib/legacy-v3.js

@@ -1,20 +0,0 @@
-import { readFileSync } from 'node:fs'
-import { baseURLFingerprint, sessionAncestry } from './route.js'
-
-const PATTERN = /\[dsh-lcx-codex-v3-checkpoint:([0-9a-f-]{36})\]/iu
-function textOf(message) { return (message?.content ?? []).filter((b) => b?.type === 'text').map((b) => b.text).join('') }
-export function legacyV3Id(message) { return textOf(message).match(PATTERN)?.[1]?.toLowerCase() }
-export function loadLegacyRecord(file, id) {
-  if (!file || !id) return undefined
-  try {
-    const parsed = JSON.parse(readFileSync(file, 'utf8'))
-    const record = parsed?.version === 3 ? parsed.checkpoints?.[id] : undefined
-    if (!record || record.version !== 3 || !Array.isArray(record.nativeOutput)) return undefined
-    return structuredClone(record)
-  } catch { return undefined }
-}
-export function legacyRouteCompatible(record, route, ctx) {
-  if (!record || record.provider !== route.provider || record.model !== route.model) return false
-  if (record.baseURLFingerprint !== baseURLFingerprint(route.baseURL)) return false
-  return sessionAncestry(ctx, route.sessionId).includes(record.sourceSessionId ?? record.lineageId)
-}

package/lib/responses-replay.js

@@ -1,68 +0,0 @@
-import { mergeFeatureHeader } from './compact-v2.js'
-import { resolvePiResponsesModel } from './dsh-responses.js'
-import { buildResponsesBody } from './responses-request.js'
-import { streamResponsesRequest } from './responses-stream.js'
-
-/**
- * Backward-compatible body helper retained for protocol tests and downstream imports.
- * Production managed replay already carries the DSH system prompt in canonical input;
- * `system` remains only as a low-level compatibility field.
- */
-export function replayBody({ model, modelDescriptor, input, system, tools, promptCacheKey, promptCacheRetention, cacheRetention, reasoningEffort, temperature, maxTokens }) {
-  const descriptor = modelDescriptor ?? resolvePiResponsesModel({
-    route: { provider: 'lcx', model, baseURL: '' },
-    model: { id: model, provider: 'lcx', baseUrl: '', api: 'openai-responses', reasoning: true, input: ['text'] },
-  })
-  const body = buildResponsesBody({
-    model: descriptor,
-    input,
-    instructions: system,
-    tools,
-    promptCacheKey,
-    promptCacheRetention,
-    cacheRetention: cacheRetention ?? (promptCacheKey ? (promptCacheRetention ? 'long' : 'short') : 'none'),
-    reasoningEffort,
-    temperature,
-    maxTokens,
-  })
-  // Opaque Native replay stays on the Remote V2 wire contract while sharing the standard builder.
-  body.tool_choice = 'auto'
-  body.parallel_tool_calls = true
-  return body
-}
-
-/**
- * Compatibility wrapper: Native replay now uses the same builder/transport/parser as ordinary turns.
- */
-export async function* requestNativeReplay({ baseURL, provider, model, modelDescriptor, input, system, tools, promptCacheKey, promptCacheRetention, cacheRetention, reasoningEffort, temperature, maxTokens, grammarToolInputProperties, headers, signal, timeoutMs, maxAttempts = 1, maxResponseBytes }) {
-  const descriptor = modelDescriptor ?? resolvePiResponsesModel({
-    route: { provider, model, baseURL },
-    model: { id: model, provider, baseUrl: baseURL, api: 'openai-responses', reasoning: true, input: ['text'] },
-  })
-  const body = replayBody({
-    model,
-    modelDescriptor: descriptor,
-    input,
-    system,
-    tools,
-    promptCacheKey,
-    promptCacheRetention,
-    cacheRetention,
-    reasoningEffort,
-    temperature,
-    maxTokens,
-  })
-  yield* streamResponsesRequest({
-    baseURL,
-    provider,
-    model,
-    piModel: descriptor,
-    body,
-    grammarToolInputProperties,
-    headers: mergeFeatureHeader(headers),
-    signal,
-    timeoutMs,
-    maxAttempts,
-    maxResponseBytes,
-  })
-}

package/scripts/probe-alpha.mjs

@@ -1,43 +0,0 @@
-import { readFileSync } from 'node:fs'
-import { homedir } from 'node:os'
-import { join } from 'node:path'
-import { randomUUID } from 'node:crypto'
-import { ALPHA_SCHEMA_FINGERPRINT, buildAlphaSearchBody, normalizeAlphaSearchArgs, parseAlphaSearchResponse, probeAlphaCapabilities } from '../lib/web-search-alpha.js'
-import { AlphaCapabilityStore, alphaCapabilityFingerprint } from '../lib/web-search-capability.js'
-
-const baseURL = String(process.env.LCX_BASE_URL ?? 'https://api.lcxbot.com/v1').replace(/\/+$/u, '')
-const model = String(process.env.LCX_MODEL ?? '').trim()
-const provider = String(process.env.LCX_PROVIDER ?? 'lcx').trim()
-const profile = String(process.env.LCX_ALPHA_PROFILE ?? '')
-const group = String(process.env.LCX_ALPHA_GROUP ?? '')
-const keyFile = String(process.env.LCX_API_KEY_FILE ?? '').trim()
-if (!model) throw new Error('Set LCX_MODEL to the exact GPT model id')
-if (!keyFile) throw new Error('Set LCX_API_KEY_FILE to a local file containing the bearer key')
-const apiKey = readFileSync(keyFile, 'utf8').trim()
-if (!apiKey) throw new Error('LCX_API_KEY_FILE is empty')
-const storeFile = process.env.LCX_ALPHA_CAPABILITY_PATH ?? join(process.env.DSH_HOME ?? join(homedir(), '.dsh'), 'storages', 'lcx-codex', 'web-alpha-capabilities.json')
-const sessionId = `lcx-probe-${randomUUID()}`
-
-async function invoke(rawArgs) {
-  const args = normalizeAlphaSearchArgs(rawArgs)
-  const requestId = randomUUID()
-  const response = await fetch(`${baseURL}/alpha/search`, {
-    method: 'POST',
-    headers: { 'content-type': 'application/json', authorization: `Bearer ${apiKey}`, 'x-client-request-id': requestId, 'session-id': sessionId },
-    body: JSON.stringify(buildAlphaSearchBody(args, model, sessionId, true, 2500)),
-  })
-  const text = await response.text()
-  let body
-  try { body = text ? JSON.parse(text) : {} } catch { body = { output: text, results: [] } }
-  if (!response.ok) { const e = new Error(body?.error?.message ?? body?.message ?? `HTTP ${response.status}`); e.status = response.status; throw e }
-  return parseAlphaSearchResponse(body, { action: args.action, capability: 'command-capable', requestId })
-}
-
-const structured = process.env.LCX_ALPHA_PROBE_STRUCTURED === '1'
-const actionProbes = structured ? {
-  image_query: { query: 'OpenAI logo' }, finance: { ticker: 'MSFT', assetType: 'equity', market: 'USA' }, weather: { location: 'San Francisco, CA' }, sports: { fn: 'standings', league: 'nba' }, time: { utcOffset: '+00:00' },
-} : {}
-const result = await probeAlphaCapabilities({ invoke, schemaFingerprint: ALPHA_SCHEMA_FINGERPRINT, trustedNativeProvenance: process.env.LCX_ALPHA_TRUST_NATIVE === '1', actionProbes, clickProbeRef: process.env.LCX_ALPHA_CLICK_PROBE_REF, screenshotProbeRef: process.env.LCX_ALPHA_SCREENSHOT_PROBE_REF })
-const fingerprint = alphaCapabilityFingerprint({ baseURL, provider, model, profile, group, schemaFingerprint: ALPHA_SCHEMA_FINGERPRINT })
-new AlphaCapabilityStore(storeFile).put(fingerprint, result)
-console.log(JSON.stringify({ fingerprint, storeFile, classification: result.classification, actions: result.actions, probedAt: result.probedAt, provenance: result.provenance }, null, 2))