Skip to content

Releases: BYK/loreai

0.24.1

22 May 19:16
Immutable release. Only release title and notes can be modified.

Choose a tag to compare

Bug Fixes 🐛

  • Retry with token-level truncation on ONNX OOM in embedding worker by @BYK in #457

0.24.0

21 May 19:44
Immutable release. Only release title and notes can be modified.

Choose a tag to compare

New Features ✨

  • Workspace discovery for monorepo support by @BYK in #451

Bug Fixes 🐛

  • (ci) Write Craft publish-state file so retries skip already-published targets by @BYK in #449
  • Detect and recover from stale OAuth credentials in background workers by @BYK in #454
  • Recover from prompt overflow errors and add safety margins by @BYK in #452

Internal Changes 🔧

  • (publish) Finalize Sentry release after publishing by @BYK in #450

0.23.0

21 May 11:09
Immutable release. Only release title and notes can be modified.

Choose a tag to compare

New Features ✨

Cch

  • Add seeds for Claude Code up to 2.1.146 by @github-actions in #436
  • Add seeds for Claude Code up to 2.1.145 by @github-actions in #412
  • Add seeds for Claude Code up to 2.1.144 by @github-actions in #379

Ui

  • Visual cost bars, savings hero, and daily trend chart by @BYK in #418
  • Chat-bubble conversation view for session page by @BYK in #398

Other

  • Migrate eval system to vitest-evals by @BYK in #441
  • Allow multiple recall tool calls per request (multi-turn recall) by @BYK in #404
  • Improve recall tool description + add cross-session cue eval scenarios by @BYK in #396
  • Inflate + replay compatibility — skip filler turns in gateway replay by @BYK in #393
  • Wire --inflate flag into eval CLI for 400K token scenario testing by @BYK in #386
  • Lightweight cross-session action tag context for curator by @BYK in #385
  • Scenario inflator for 400K token eval scenarios by @BYK in #384
  • Add local LLM provider support (vllm, ollama, llama.cpp, etc.) by @BYK in #383
  • Add temperature parameter to LLMClient interface by @BYK in #382
  • Lower echo threshold + clustering for broader behavioral pattern detection by @BYK in #377
  • Action tagging in distillation for implicit behavioral pattern detection by @BYK in #376
  • Vector similarity-based behavioral pattern detection by @BYK in #375
  • Add session recording, replay, and scenario filtering to eval harness by @BYK in #374
  • Add comprehensive eval suite for Lore's five key dimensions by @BYK in #369
  • Tool-call-aware cache warming + /lore:warm:* commands + UI controls by @BYK in #370

Bug Fixes 🐛

  • (cache-warmer) Tighten warming heuristics to reduce net negative spend by @BYK in #429
  • (eval) Build realistic Lore context for QA questions by @BYK in #414
  • (ui) Daily cost trend chart not rendering bars by @BYK in #426
  • 1h TTL cache write pricing and dynamic savings/overhead wording by @BYK in #445
  • Resolve Sentry issues — JSONC config, ENOENT guard, Node.js port fallback, transient error filtering by @BYK in #443
  • Use compaction as primary baseline, fix threshold, update marketing by @BYK in #437
  • Context health note — signal omitted details instead of false assurance by @BYK in #434
  • Scripted interceptor for eval replay + distillation prompt tuning by @BYK in #433
  • Downweight knowledge in recall when session content exists by @BYK in #432
  • Store scripted assistant content in eval + session-affinity recall boost by @BYK in #431
  • Add amnesia mode and x-lore-no-store to prevent QA contamination in recall by @BYK in #430
  • Prevent tool_use/tool_result mismatch at gradient prefix/raw boundary (#424) by @BYK in #428
  • Resume sessions across client restarts instead of orphaning them by @BYK in #427
  • Create dev shims in gateway build so workspace consumers resolve bun exports by @BYK in #425
  • Drop callCount from dedup tests — still flaky after #419 by @BYK in #421
  • Use promise identity in dedup tests to eliminate cross-test flakiness by @BYK in #419
  • Build Bun-native ESM bundle for OpenCode plugin by @BYK in #415
  • Remove distilled=0 filter from temporal BM25 search by @BYK in #413
  • Add deepseek- dash-prefix routing to UPSTREAM_ROUTES by @rgutzen in #406
  • Add anti-imitation instruction to recall tool description by @BYK in #411
  • Inject X-Lore-Project header to eliminate cwd fallback warning by @BYK in #408
  • Code block alignment in Simplicity & Migration sections by @BYK in #407
  • Background worker rate-limit resilience for Claude Max by @BYK in #397
  • QA prompt instructs LLM to use recall tool by @BYK in #400
  • Pin user assertions in distillation to prevent loss in long sessions by @BYK in #394
  • Fall back to upstream when compaction summary generation fails by @BYK in #392
  • Pass local:true when starting gateway in-process to prevent hosted mode by @BYK in #390
  • Curator uses distilled observations when all messages are distilled by @BYK in #389
  • Judge prompt should not penalize extra correct information by @BYK in #388
  • Remove distilled=0 filter from temporal vector search by @BYK in #387
  • Preserve temporal embeddings on distillation + expose source IDs in recall by @BYK in #380
  • Improve distillation detail preservation for code-specific artifacts by @BYK in #378
  • Warmup cooldown timing + rising cost threshold + accurate TTL pricing by @BYK in #373
  • Improve curator preference detection and evolution handling by @BYK in #372

Documentation 📚

  • Replace waitlist form with Loops.so integration by @BYK in #447
  • Hero stat 13/20 → 2.6x total recall by @BYK in #444
  • Update website copy — sessions lasting days, crystal-clear memory by @BYK in #442
  • Add releasing instructions to AGENTS.md by @BYK in #438
  • Update eval results — context retention 3.9→4.6, +77% vs tail-window by @BYK in #435
  • Update website and README with context retention eval results by @BYK in #420
  • Update website hero stats with 400K eval results by @BYK in #405
  • Update README with 400K token eval results by @BYK in #403

Internal Changes 🔧

  • (deps) Bump pytest from 8.2.2 to 9.0.3 in /packages/core/eval/fixtures/projects/python-api in the pip group across 1 directory by @dependabot in #371
  • Disable eval workflow on PRs by @BYK in #446
  • Improve distillation detail retention at 400K+ token sessions by @BYK in #423
  • Remove deprecated eval infra and update docs with new eval results by @BYK in #422

Other

  • eval: add 2.3M-token mega-session scenario — Lore 4.0 vs Compaction 2.4 (+70%) by @BYK in #440

0.22.0

16 May 21:54
Immutable release. Only release title and notes can be modified.

Choose a tag to compare

New Features ✨

  • (cch) Add seeds for Claude Code up to 2.1.143 by @github-actions in #356
  • (website) Add unified architecture section — context + memory as one pipeline by @BYK in 0a6587a9
  • Rewrite README to align with new website messaging by @BYK in #366
  • Marketing rewrite + product improvements inspired by Stellman/Orosz articles by @BYK in #362

Bug Fixes 🐛

  • Expand OpenCode plugin GATEWAY_PROVIDERS to route all proxiable providers through gateway by @BYK in #365
  • Resolve high priority bugs from quality playbook phase 3 by @BYK in #364
  • Stop attributing all turns after warmup as cache warming savings by @BYK in #363
  • Add PNG/ICO favicon fallbacks for Firefox Android by @BYK in 9d31559e
  • Add favicon to website matching dashboard by @BYK in 8e34d32d

0.21.0

16 May 13:32
Immutable release. Only release title and notes can be modified.

Choose a tag to compare

New Features ✨

  • Display subagent sessions as collapsible tree with cost roll-up by @BYK in #360
  • Persist table sort order to localStorage by @BYK in #354

Bug Fixes 🐛

  • Correct cache bust detection causing false unsustainable warnings by @BYK in #359
  • Improve LTM preference injection — skip relevance scoring, dedicated budget, meaningful confidence by @BYK in #358
  • Forward anthropic-beta header in cache-warmer to support beta-gated body fields by @BYK in #357
  • Prevent splitSegments infinite recursion and add global background work throttling by @BYK in #355

0.20.2

15 May 18:45
Immutable release. Only release title and notes can be modified.

Choose a tag to compare

Bug Fixes 🐛

  • Review follow-ups for unsustainable warning PR by @BYK in #352
  • Surface unsustainable warning on all gradient return paths by @BYK in #351

Internal Changes 🔧

  • Replace EMA-driven context cap with tier-based cost-aware decisions by @BYK in #348

Other

0.20.1

15 May 18:04
Immutable release. Only release title and notes can be modified.
0.20.1
fb00cc2

Choose a tag to compare

Internal Changes 🔧

  • Replace EMA-driven context cap with tier-based cost-aware decisions by @BYK in #348

0.20.0

15 May 15:50
Immutable release. Only release title and notes can be modified.

Choose a tag to compare

New Features ✨

  • (cch) Add seeds for Claude Code up to 2.1.142 by @github-actions in #329
  • REST API for remote data management, recall, and import by @BYK in #337
  • Lore start defaults to hosted mode, opt out with --local by @BYK in #334
  • LORE_HOSTED_MODE — disable FS operations on client-controlled paths by @BYK in #333
  • X-Lore-Git-Remote header and LORE_REMOTE_URL for remote gateway project grouping by @BYK in #331

Bug Fixes 🐛

  • Resolve pi build rootDir error by adding types export to core by @BYK in #346
  • Prevent Sentry noise from embedding failures and fix surrogate pair truncation by @BYK in #344
  • Pre-truncate long texts to prevent ONNX OOM and report embedding errors to Sentry by @BYK in #343
  • Break Layer 4 cache hit rate death spiral by @BYK in #341
  • Exempt sub-agent sessions from cache warming by @BYK in #340
  • Harden remote import error handling and clean up code by @BYK in #339
  • Import dedup checks remote DB when LORE_REMOTE_URL is set by @BYK in #338
  • Stop merging sub-agent turns into parent session by @BYK in #335
  • Use text blocks in recall follow-up to prevent tool leaking to client by @BYK in #332
  • Keep recall tool in follow-up request to prevent API rejection by @BYK in #330

Internal Changes 🔧

  • (ui) Unify Per Session and Live Sessions tables into shared renderer by @BYK in #336
  • Extract ltmDiffThreshold() helper to deduplicate step 6 and 7b by @BYK in #342

0.19.0

14 May 21:19
Immutable release. Only release title and notes can be modified.

Choose a tag to compare

New Features ✨

  • (cch) Add seeds for Claude Code up to 2.1.141 by @github-actions in #297
  • Adaptive dedup threshold with automatic and manual calibration by @BYK in #321
  • Add logo favicon to web dashboard pages by @BYK in f94c4cf6
  • Show live/total breakdown in dashboard and cost pills by @BYK in #317
  • 3-block system prompt for cache-efficient LTM injection by @BYK in #311
  • Defer context-dependent LTM to turn 2, add hybrid vector+FTS5 scoring by @BYK in #310
  • Add OpenAI Batch API support for worker calls by @BYK in #309
  • Re-run forSession() for fresh LTM on Layer 4 emergency reset by @BYK in #305
  • Crash-safe state persistence via periodic flush by @BYK in #303

Bug Fixes 🐛

  • Redirect GET / to /ui dashboard by @BYK in #327
  • Preserve thinking blocks in recall follow-up to prevent API rejection by @BYK in #326
  • Resolve embedding worker crash in CJS bundle (LOREAI-GATEWAY-D/E) by @BYK in #325
  • Pass project_id (not path) to projectName() on warming costs table by @BYK in #324
  • Prevent fatal crash when startServer() called with missing hosts by @BYK in #322
  • Remove text from logo SVG, keep icon only by @BYK in 42e27682
  • Use icon-only logo (no text) for favicon by @BYK in ecab9388
  • Prevent test data from leaking into production DB by @BYK in #320
  • Use export for env vars in install script examples by @BYK in #319
  • Key global histograms by project_id instead of directory path by @BYK in #318
  • Tidy dashboard and cost intelligence pills for scope consistency by @BYK in #316
  • Pass --since tag to craft changelog for shallow clone compatibility by @BYK in #315
  • Show craft stderr on nightly version failure for debugging by @BYK in #314
  • Compute nightly version from next semver bump instead of current release by @BYK in #313
  • Address review findings from 3-block system prompt PR by @BYK in #312
  • Strip recall marker text from temporal storage to prevent FTS echo by @BYK in #306
  • Pre-populate headerSessionIndex from DB on startup by @BYK in #304
  • Prevent layer-4 stickiness trap and refactor compression stages by @BYK in #300
  • Add concurrency guards and persist volatile session state across restarts by @BYK in #298

Documentation 📚

  • Recover install.sh curl copy/paste block on landing page by @BYK in eb671df7
  • Change Lore to Lore.AI and adjust scaling by @sylncnr in #308
  • Update landing page branding and header layout by @sylncnr in #307

0.18.0

13 May 17:27
Immutable release. Only release title and notes can be modified.

Choose a tag to compare

New Features ✨

  • (cch) Add seeds for Claude Code up to 2.1.140 by @github-actions in #283
  • (cli) Forward extra arguments to launched agent by @BYK in #284
  • (core) Detect repeated user instructions for LTM curation by @BYK in #269
  • (gateway) Add OpenAI Responses API wire protocol support by @BYK in #263
  • (warming) Replace per-window survival with commitment-based warming model by @BYK in #280
  • Add embedding-based dedup and 'lore data reindex' command by @BYK in #288
  • Migrate embeddings from fastembed/BGE Small to transformers.js/Nomic v1.5 by @BYK in #287
  • Add nightly build support to install script by @BYK in #275
  • Change default port from 6969 to 3207 with fallback chain by @BYK in #273
  • Decouple Pi plugin from @loreai/core via gateway /v1/compact endpoint by @BYK in #271
  • Boost vector search weight in RRF for semantic queries by @BYK in #260
  • Rename data directory from opencode-lore to lore with auto-migration by @BYK in #259
  • Add curator observability and fix pattern extraction false positives by @BYK in #257
  • Add cache warming dashboard page and session section by @BYK in #256
  • Show dashboard URL on startup and add lore logs command by @BYK in #255
  • Improve meta-request detection to prevent distilling title/summary conversations by @BYK in #253
  • Add conversation import system for extracting knowledge from prior AI agent history by @BYK in #252
  • Add user-level knowledge page to dashboard by @BYK in #248

Bug Fixes 🐛

  • (cch) Make seed check CI version-agnostic and extract all missing versions by @BYK in #279
  • Include global entries in dedup scan by @BYK in #295
  • Tune embedding dedup threshold to 0.935 by @BYK in #293
  • Raise embedding dedup threshold from 0.85 to 0.92 by @BYK in #291
  • Use token-budget batching to prevent OOM on long texts by @BYK in #290
  • Prevent OOM in embedding backfill with long texts by @BYK in #289
  • Add word-overlap dedup for knowledge entries by @BYK in #286
  • Raise default maxListeners to suppress benign gz stream warning by @BYK in #285
  • Deduplicate concurrent models.dev fetch requests by @BYK in #282
  • Improve lore import UX — stdout logging, prompt ordering, only-ask-once by @BYK in #278
  • Upgrade double-lock, flag parsing, and message ordering by @BYK in #277
  • Cache inferred project path in session state to avoid cwd fallback by @BYK in #276
  • Correct cost intelligence accuracy and add overhead breakdown by @BYK in #272
  • Filter automated turns from cache warming gap histograms by @BYK in #270
  • Streaming robustness, RRF list priority, and CLI suggestion quality by @BYK in #268
  • Quality improvements across cache, recall, streaming, and CLI by @BYK in #267
  • Prevent post-idle compaction from busting /keep sessions on 5m TTL by @BYK in #266
  • Use persisted real worker costs in historical estimates instead of heuristic by @BYK in #265
  • Remove stale project path caching that caused cross-project misattribution by @BYK in #262
  • Filter distilled messages from temporal FTS search and purge worker boilerplate by @BYK in #261
  • Skip post-idle compaction when cache warmer kept the cache warm by @BYK in #258
  • Prevent subagent turns from inflating session messageCount by @BYK in #254
  • Write informational CLI messages to stdout instead of stderr by @BYK in #249
  • Import knowledge from .lore.md on file change and new session start by @BYK in #247
  • Reuse existing gateway instead of crashing on EADDRINUSE by @BYK in #246
  • Avoid Bun NAPI teardown crash on gateway exit by @BYK in #245

Documentation 📚

  • Comprehensive mobile layout and grid overflow fixes by @sylncnr in #250

Internal Changes 🔧

  • (ci) Upgrade GitHub Actions to Node.js 24 compatible versions by @BYK in #281

Other

  • cleanup: remove "gateway mode" language by @BYK in #274