Skip to content

CHANGELOG.md

Latest commit

 

History

History
922 lines (716 loc) · 39.5 KB

CHANGELOG.md

File metadata and controls

922 lines (716 loc) · 39.5 KB

Changelog

All notable changes to LibreFang will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[0.6.6] - 2026-03-20

Fixed

  • Use 'file' instead of 'dockerfile' in docker/build-push-action@v7 (#1298) (@houko)

[0.6.5] - 2026-03-20

Added

  • Auto-initialize vault during librefang init (#1206) (@houko)
  • Add token consumption metadata and reduce default hands (#1215) (@houko)
  • Add image pipeline and subprocess management (#1223) (@f-liva)
  • Add Qwen Code CLI as LLM provider (#1224) (@f-liva)
  • Align init defaults with OpenRouter Stepfun (#1262) (@houko)
  • Replace all icons with new LibreFang branding (#1263) (@f-liva)
  • Fix shell (#1270) (@houko)

Fixed

  • Decrypt encrypted webhook payloads (#1208) (@TechWizard9999)
  • Bootstrap context engine during startup (#1209) (@TechWizard9999)
  • Support target ids in channel test (#1210) (@TechWizard9999)
  • Make shell installer POSIX-compatible for Linux (#1226) (@houko)
  • Web deployment issues (#1236) (@houko)
  • Web deployment issues (#1237) (@houko)
  • Create web/public/assets directory (#1238) (@houko)
  • Web deployment and CI fixes (#1239) (@houko)
  • Web deployment and CI fixes (#1243) (@houko)
  • Repair install smoke script and drop update-star-history workflow (#1255) (@houko)
  • Use web/public installer source and harden curl|sh install flow (#1259) (@houko)
  • Address code-scanning path-injection findings (follow-up) (#1260) (@houko)
  • Initialize rustls crypto provider for TLS connections (#1294) (@houko)

Documentation

  • Update star history (#1197) (@houko)
  • Update star history (#1198) (@houko)
  • Update star history (#1199) (@houko)
  • Update star history (#1200) (@houko)
  • Update star history (#1201) (@houko)
  • Update star history (#1202) (@houko)
  • Update star history (#1203) (@houko)
  • Update star history (#1213) (@houko)
  • Update contributors (#1214) (@houko)
  • Update star history (#1225) (@houko)
  • Update star history (#1228) (@houko)
  • Add SDK usage examples to all README files (#1229) (@houko)
  • Update star history (#1231) (@houko)
  • Update star history (#1232) (@houko)
  • Update star history (#1235) (@houko)
  • Update contributors (#1240) (@app/github-actions)
  • Update star history (#1242) (@app/github-actions)

Maintenance

  • Tidy repo structure (#1211) (@houko)
  • Use .nvmrc for web Node.js version and fix Dockerfile path (#1234) (@houko)
  • Bump actions/setup-node from 4 to 6 (#1280) (@app/dependabot)
  • Bump actions/download-artifact from 4 to 8 (#1281) (@app/dependabot)
  • Bump pnpm/action-setup from 4 to 5 (#1282) (@app/dependabot)
  • Bump actions/labeler from 5 to 6 (#1283) (@app/dependabot)
  • Bump zip from 8.2.0 to 8.3.0 (#1284) (@app/dependabot)
  • Bump jsonwebtoken from 9.3.1 to 10.3.0 (#1285) (@app/dependabot)
  • Bump tracing-subscriber from 0.3.22 to 0.3.23 (#1286) (@app/dependabot)
  • Bump rusqlite from 0.38.0 to 0.39.0 (#1287) (@app/dependabot)
  • Bump rumqttc from 0.24.0 to 0.25.1 (#1288) (@app/dependabot)
  • Bump tokio-tungstenite from 0.28.0 to 0.29.0 (#1289) (@app/dependabot)
  • Bump criterion from 0.5.1 to 0.8.2 (#1290) (@app/dependabot)
  • Bump rand from 0.8.5 to 0.9.2 (#1291) (@app/dependabot)
  • Bump toml_edit from 0.25.4+spec-1.1.0 to 0.25.5+spec-1.1.0 (#1292) (@app/dependabot)

Other

  • Clean skills (#1212) (@houko)
  • Fix/webui chat input line break failed (#1245) (@aimlyo)

[v0.6.4-20260320] - 2026-03-20

Added

  • Add image pipeline and subprocess management (#1223) (@f-liva)
  • Add Qwen Code CLI as LLM provider (#1224) (@f-liva)
  • Align init defaults with OpenRouter Stepfun (#1262) (@houko)
  • Replace all icons with new LibreFang branding (#1263) (@f-liva)

Fixed

  • Web deployment issues (#1236) (@houko)
  • Web deployment issues (#1237) (@houko)
  • Create web/public/assets directory (#1238) (@houko)
  • Web deployment and CI fixes (#1239) (@houko)
  • Web deployment and CI fixes (#1243) (@houko)
  • Repair install smoke script and drop update-star-history workflow (#1255) (@houko)
  • Use web/public installer source and harden curl|sh install flow (#1259) (@houko)
  • Address code-scanning path-injection findings (follow-up) (#1260) (@houko)

Documentation

  • Update star history (#1228) (@houko)
  • Add SDK usage examples to all README files (#1229) (@houko)

Maintenance

  • Use .nvmrc for web Node.js version and fix Dockerfile path (#1234) (@houko)

Other

  • Fix/webui chat input line break failed (#1245) (@aimlyo)

[0.6.4] - 2026-03-20

Added

  • Add image pipeline and subprocess management (#1223) (@f-liva)
  • Add Qwen Code CLI as LLM provider (#1224) (@f-liva)
  • Align init defaults with OpenRouter Stepfun (#1262) (@houko)
  • Replace all icons with new LibreFang branding (#1263) (@f-liva)

Fixed

  • Web deployment issues (#1236) (@houko)
  • Web deployment issues (#1237) (@houko)
  • Create web/public/assets directory (#1238) (@houko)
  • Web deployment and CI fixes (#1239) (@houko)
  • Web deployment and CI fixes (#1243) (@houko)
  • Repair install smoke script and drop update-star-history workflow (#1255) (@houko)
  • Use web/public installer source and harden curl|sh install flow (#1259) (@houko)
  • Address code-scanning path-injection findings (follow-up) (#1260) (@houko)

Documentation

  • Add SDK usage examples to all README files (#1229) (@houko)
  • Update contributors (#1240) (@app/github-actions)
  • Update contributors and star history (#1244) (@app/github-actions)
  • Update contributors and star history (#1246) (@app/github-actions)
  • Update contributors and star history (#1247) (@app/github-actions)
  • Update contributors and star history (#1248) (@app/github-actions)
  • Update contributors and star history (#1250) (@app/github-actions)
  • Update contributors and star history (#1251) (@app/github-actions)
  • Update contributors and star history (#1253) (@app/github-actions)
  • Update contributors and star history (#1256) (@app/github-actions)
  • Update contributors and star history (#1257) (@app/github-actions)
  • Update contributors and star history (#1258) (@app/github-actions)
  • Update contributors and star history (#1261) (@app/github-actions)
  • Update contributors and star history (#1264) (@app/github-actions)
  • Update contributors and star history (#1265) (@app/github-actions)
  • Update contributors and star history (#1267) (@app/github-actions)

Maintenance

  • Use .nvmrc for web Node.js version and fix Dockerfile path (#1234) (@houko)

Other

  • Fix/webui chat input line break failed (#1245) (@aimlyo)

[0.6.3] - 2026-03-19

Added

  • Auto-initialize vault during librefang init (#1206) (@houko)
  • Add token consumption metadata and reduce default hands (#1215) (@houko)

Fixed

  • Decrypt encrypted webhook payloads (#1208) (@TechWizard9999)
  • Bootstrap context engine during startup (#1209) (@TechWizard9999)
  • Support target ids in channel test (#1210) (@TechWizard9999)
  • Make shell installer POSIX-compatible for Linux (#1226) (@houko)

Documentation

  • Update contributors (#1214) (@houko)

Maintenance

  • Tidy repo structure (#1211) (@houko)

Other

  • Clean skills (#1212) (@houko)

[0.6.2] - 2026-03-19

Fixed

  • Prevent provider appearing in multiple tier groups (#1190) (@SenZhangAI)
  • Resolve 17 compilation errors breaking CI (#1193) (@houko)

[0.6.1] - 2026-03-18

Added

  • Graceful degradation when no LLM provider configured (#1185) (@SenZhangAI)

Fixed

  • Remove markdown fence wrapper from dev.to articles (#1167) (@houko)
  • Resolve secret scanning alert for MongoDB example URI (#1168) (@houko)
  • Handle paginated response in agents list and chat resolver (#1169) (@houko)
  • Resolve agent names to UUIDs in message and kill commands (#1170) (@houko)
  • Return 409 Conflict when spawning duplicate agent (#1171) (@houko)
  • Parse model aliases from API response correctly (#1172) (@houko)
  • Include last_active in agent detail endpoint (#1173) (@houko)
  • Parse wrapped API responses in CLI table views (#1175) (@houko)
  • Resolve agent names in trigger, cron, and webhook commands (#1176) (@houko)
  • Complete dashboard i18n translation coverage (#1177) (@houko)
  • Webhook CLI commands use wrong API endpoints (#1178) (@houko)
  • A2A agent card uses service config instead of random agent (#1179) (@houko)
  • Budget PUT accepts GET response field names for read-modify-write (#1182) (@houko)
  • Models set sends wrong field name to config/set API (#1183) (@houko)
  • Cron create returns proper JSON instead of stringified blob (#1184) (@houko)
  • CLI cron list reads nested schedule/action fields (#1186) (@houko)
  • Triggers list returns wrapped object for consistency (#1187) (@houko)
  • Include system_prompt in GET /api/agents/:id response (#1188) (@houko)

Maintenance

  • Fix rustfmt in a2a_agent_card handler (#1181) (@houko)

[0.6.0] - 2026-03-18

Added

  • Add filtering, pagination and sorting to agent list endpoint (#399) (@houko)
  • Add HTTP proxy support for all outbound connections (#415) (@houko)
  • Auto-register local workflow definitions at daemon startup (#418) (@houko)
  • Add multimedia support for Telegram and Discord channels (#422) (@houko)
  • Add Telegram streaming output with progressive message updates (#423) (@houko)
  • Add NVIDIA NIM as dedicated LLM provider (#428) (@houko)
  • Add MQTT pub/sub channel adapter for IoT integration (#430) (@houko)
  • Add workflow trigger support to cron jobs (#431) (@houko)
  • Add hierarchical Goals system with REST API and dashboard UI (#434) (@houko)
  • Bundle Python and Node.js runtimes in Docker image (#334) (#436) (@houko)
  • Add Vertex AI driver with OAuth2 authentication (#448) (@houko)
  • Add GET /api/providers/:name endpoint (#1090) (@houko)
  • Add GET /api/workflows/:id endpoint (#1091) (@houko)
  • Add GET /api/channels/:name endpoint (#1092) (@houko)
  • Add GET /api/cron/jobs/:id endpoint (#1093) (@houko)
  • Add GET /api/mcp_servers/:name endpoint (#1094) (@houko)
  • Add PUT/DELETE /api/workflows/:id endpoints (#1095) (@houko)
  • Add DELETE /api/agents/:id/files/:filename endpoint (#1097) (@houko)
  • Add Workflow variant to CronAction for cron-triggered workflows (#1102) (@houko)
  • Propagate sender identity from channels to agent context (#1105) (@houko)
  • Auto-register local workflow definitions at daemon startup (fixes #382) (#1107) (@houko)
  • Implement mem0-style proactive memory system (#1111) (@houko)
  • Web search key rotation, data-driven hand routing, and health-aware LLM fallback (#1127) (@houko)
  • Improve context engine accuracy and resilience (#1146) (@houko)
  • Add context engine plugin management system (#1152) (@houko)
  • Support multiple custom plugin registries (#1154) (@houko)

Fixed

  • Extract thread_ts from Slack events for thread replies (#1099) (@houko)
  • Add mime_type to ChannelContent::Image for correct vision handling (#1100) (@houko)
  • Use SHA-256 for Nostr pubkey derivation instead of DefaultHasher (#1101) (@houko)
  • Prevent silent message dropping in Telegram dispatch (#1103) (@houko)
  • Handle thought chunks in Gemini streaming for thinking models (#1104) (@houko)
  • Don't break streaming bridge on intermediate ContentComplete (#1126) (@houko)
  • Fall back to bundled Mozilla CA roots when system certs unavailable (#1142) (@houko)
  • Upstream parity — 10 bug fixes from release comparison (#1143) (@houko)
  • Resolve clippy warnings, test failures, and add agent list validation (#1162) (@houko)

Documentation

  • Slim down README (#1124) (@houko)

Maintenance

  • Skip version bump PRs in changelog generation (#1123) (@houko)
  • Bump setup-python v5→v6 and create-pull-request v7→v8 (#1161) (@houko)

[0.5.7] - 2026-03-18

Added

  • Add include_skills and include_tools flags to agent clone API (#366) (@houko)
  • Add event webhooks API for system events (#394) (@houko)
  • Add task queue management API (#395) (@houko)
  • Add agent monitoring and metrics API (#396) (@houko)
  • Add API input validation middleware (#398) (@houko)
  • Add webhooks management API (#400) (@houko)
  • Add 6 new bundled hand templates (#413) (@houko)
  • Add multi-agent orchestration foundation (#323) (#437) (@houko)
  • Add Feishu interactive card approval for agent permission requests (#439) (@houko)
  • Add multi-token fallback with transparent quota rotation (#441) (@houko)
  • Add JWT/service account auth to Google Chat adapter (#443) (@houko)
  • Auto-post release article to GitHub Discussions (#582) (@houko)
  • Add GET /api/integrations/:id endpoint (#1088) (@houko)
  • Add GET /api/approvals/:id endpoint (#1089) (@houko)
  • Add GET /api/sessions/:id endpoint (#1096) (@houko)
  • Telegram streaming output support with progressive typing effect (fixes #317) (#1109) (@houko)

Fixed

  • YAML syntax errors + auto-post release to GitHub Discussions (#567) (@houko)
  • Inherit parent env in MCP subprocess instead of clearing (#1098) (@houko)
  • Send empty object instead of null for parameterless tool calls (fixes #918) (#1108) (@houko)
  • Add missing TokenUsage fields in token rotation test (#1114) (@houko)

[0.5.6] - 2026-03-17

Added

  • Add multi-language support for CLI and API error messages (#449) (@houko)
  • Add truncation and metadata for Telegram reply-to-message (#560) (@SenZhangAI)

Fixed

  • SDK publish fixes + Bluesky notification + auto Dev.to article (#562) (@houko)
  • YAML syntax error in Bluesky notification workflow (#563) (@houko)
  • Add missing discovered_model_info field in ProbeResult test (#565) (@houko)

[0.5.5] - 2026-03-17

Added

  • Add Telegram reply-to-message context (#553) (@SenZhangAI)
  • Enrich Ollama model discovery with metadata (#554) (@SenZhangAI)
  • Add GET /api/peers/{id} endpoint (#557) (@SenZhangAI)

Fixed

  • Improve Telegram markdown formatting for headings, lists, code blocks and blockquotes (#405) (@houko)
  • Normalize OpenRouter model IDs to prevent 400 errors (#408) (@houko)
  • Improve python3 detection and Chromium sandbox handling for Linux (#410) (@houko)
  • Prevent Mastodon adapter from re-delivering old notifications and posting errors publicly (#411) (@houko)
  • Replace unsafe pointer mutation with OnceLock for peer_registry/peer_node (#414) (@houko)
  • Raise main_lane default concurrency from 1 to 3 (#552) (@SenZhangAI)
  • Update static linking check to match static-pie binaries (#558) (@houko)

Performance

  • Optimize channel hot-path with reduced allocations and Criterion benchmarks (#451) (@houko)

Documentation

  • Update contributors (#555) (@houko)

Maintenance

  • Auto-cancel old release runs when tag is re-pushed (#547) (@houko)

[0.5.4] - 2026-03-17

Added

  • Add bulk operations API for agents (#397) (@houko)
  • Add Z.AI and Kimi 2 model support (#409) (@houko)
  • Add static Linux binary builds with musl target (#438) (@houko)
  • Add multi-provider OAuth/OIDC authentication support (#454) (@houko)
  • Add session retention policy with automatic cleanup (#516) (@houko)
  • Add configurable message queue with concurrency settings (#517) (@houko)
  • Add multi-language SDKs (JavaScript, Python, Go, Rust) (#531) (@houko)
  • Auto-generate OpenAPI spec with utoipa (#534) (@houko)

Fixed

  • Complete vertex ai config wiring (#498) (@houko)
  • Trim message history at safe turn boundaries (#521) (@houko)
  • Add logging for X-API-Version header insertion failures (#524) (@houko)
  • Fix SDK publishing (PyPI, npm, crates.io, GHCR) (#537) (@houko)
  • Make release creation idempotent (#539) (@houko)
  • Force-push tag in release.sh to handle re-releases (#540) (@houko)
  • Use file instead of ldd to verify static linking (#541) (@houko)
  • Allow re-release to overwrite existing assets (#542) (@houko)
  • Allow desktop re-release to overwrite existing assets (#543) (@houko)
  • Make SDK publishing idempotent for re-releases (#544) (@houko)
  • Re-fetch PREV_TAG after deleting old tag in release.sh (#545) (@houko)

Changed

  • Split monolithic routes.rs into domain-specific modules (#452) (@houko)

Maintenance

  • Move binary size check from PR to release-only (#528) (@houko)
  • Split release workflow into independent parallel pipelines (#533) (@houko)

[0.5.3] - 2026-03-17

Added

  • Add bulk operations API for agents (#397) (@houko)
  • Add Z.AI and Kimi 2 model support (#409) (@houko)
  • Add static Linux binary builds with musl target (#438) (@houko)
  • Add multi-provider OAuth/OIDC authentication support (#454) (@houko)
  • Add session retention policy with automatic cleanup (#516) (@houko)
  • Add configurable message queue with concurrency settings (#517) (@houko)
  • Add multi-language SDKs (JavaScript, Python, Go, Rust) (#531) (@houko)
  • Auto-generate OpenAPI spec with utoipa (#534) (@houko)

Fixed

  • Complete vertex ai config wiring (#498) (@houko)
  • Trim message history at safe turn boundaries (#521) (@houko)
  • Add logging for X-API-Version header insertion failures (#524) (@houko)

Changed

  • Split monolithic routes.rs into domain-specific modules (#452) (@houko)

Maintenance

  • Move binary size check from PR to release-only (#528) (@houko)
  • Split release workflow into independent parallel pipelines (#533) (@houko)

[0.5.2] - 2026-03-16

Fixed

  • Auto-update contributors list from GitHub API (#512) (@houko)
  • Use local SVG for contributors with circular avatars (#513) (@houko)
  • WeCom secret env pattern + add pre-commit fmt hook (#518) (@houko)

Maintenance

  • Auto-merge release PRs after CI passes (#511) (@houko)

[0.5.1] - 2026-03-16

Fixed

  • Improve API version negotiation and local provider detection (#507) (@houko)
  • Inject vault secrets into process env at startup (#509) (@houko)

[0.5.0] - 2026-03-16

Added

  • Add GET /api/commands/:name endpoint (#369) (@houko)
  • Add recipe-assistant agent template (#393) (@houko)
  • Add Nix flake support (#412) (@houko)
  • Add Qwen Code CLI as LLM provider (#417) (@houko)
  • Add LLM provider prompt caching support (#381) (#424) (@houko)
  • Add decision trace layer for tool selection reasoning (#426) (@houko)
  • Add stable_prefix_mode for cache-friendly prompts (#427) (@houko)
  • Replace native-tls with rustls for IMAP channel (#432) (@houko)
  • Add API endpoint versioning support (#450) (@houko)
  • Generate versioned homebrew formula on release (#503) (@houko)

Fixed

  • Use default_model from config in Web UI agent creation (#402) (@houko)
  • Apply log_level from config.toml to tracing subscriber (#404) (@houko)
  • Correctly read nested tokens.id_token for Codex CLI OAuth (#406) (@houko)
  • Use deterministic UUIDs for hand agents to persist across restarts (#407) (@houko)
  • Update nix flake for nixpkgs darwin SDK migration (#491) (@houko)
  • Update nix flake for darwin SDK and crane warnings (#493) (@houko)
  • Add git to devShell and preserve user PATH (#494) (@houko)
  • Remove duplicate /api/versions route causing panic on startup (#501) (@houko)
  • Use Render API for heartbeat + release script improvements (#504) (@houko)
  • Allow re-release by replacing existing changelog entry (#505) (@houko)

Documentation

  • Improve CLI --help descriptions for all subcommands (#453) (@houko)

[0.4.7] - 2026-03-15

Added

  • Add backup and restore functionality for kernel state (#444) (@houko)
  • Add thread_id and attachments to CommsSendRequest (#469) (@TJUEZ)

Fixed

  • Resolve WhatsApp Web gateway E2EE, agent UUID, and auto-connect failures (#440) (@houko)
  • Wire thread_id and attachments in comms_send handler (#479) (@houko)
  • Include tauri.conf.json in release script git add (#482) (@houko)
  • Strip date suffix from Tauri version for Windows MSI builds (#485) (@houko)

Documentation

  • Translate getting-started.md to French (#442) (@houko)
  • Translate skill-development.md to Chinese (#447) (@houko)

[0.4.6] - 2026-03-15

Fixed

  • Enable all channel features by default and fix changelog dedup (#473) (@houko)

[0.4.5] - 2026-03-15

Added

  • Add academic-researcher agent template (#391) (@houko)
  • Add code-review-checklist prompt skill (#377) (@houko)
  • Add API endpoints for managing extensions (#372) (@houko)
  • Add memory/knowledge graph export and import API (#371) (@houko)
  • Add POST/PUT/DELETE endpoints for MCP server config management (#370) (@houko)
  • Add POST/DELETE endpoints for model aliases (#364) (@houko)
  • Add GET /api/profiles/:name endpoint (#363) (@houko)
  • Add GET /api/tools/:name endpoint (#360) (@houko)
  • Add GET /api/schedules/:id endpoint (#291) (@houko)
  • Add GET /api/a2a/agents/:id endpoint (#290) (@houko)
  • Add PUT /api/cron/jobs/:id endpoint for updating cron jobs (#289) (@houko)
  • Horizontal scroll for long commands on deploy page (#276) (@houko)
  • Add tooltip for truncated commands on deploy page (#275) (@houko)
  • Add copy buttons to install commands on deploy hub (#258) (@houko)
  • Add macOS, Linux, Windows install options to deploy hub (#257) (@houko)
  • Deploy hub with multi-platform support (#251) (@houko)
  • Add GCP free-tier deployment with Terraform (#249) (@houko)
  • Support multi-bot routing per platform (#240) (@houko)

Fixed

  • Default to all-channels (#466) (@houko)
  • Remove default-features=false to enable channel features (#465) (@houko)
  • Respect PORT env var for Railway/Render compatibility (#455) (@houko)
  • Stop syncing agent.toml versions with project release version (#375) (@houko)
  • Skip pre-release tags when finding previous version for changelog (#374) (@houko)
  • Catalog sync fails to parse remote files missing provider field (#362) (@houko)
  • Add reconnect logic to Matrix channel adapter (#361) (@houko)
  • Remove VOLUME directive from Dockerfile (#294) (@houko)
  • Render card empty due to nested anchor tags (#292) (@houko)
  • Log warnings instead of silently ignoring errors in API endpoints (#288) (@houko)
  • Add URL validation to A2A discover endpoint to prevent SSRF (#287) (@houko)
  • Validate environment variable names in channel config API (#286) (@houko)
  • Use agent_id path parameter in KV memory endpoints (#285) (@houko)
  • Use WEBSITE_REPO_TOKEN for star history workflow (#281) (@houko)
  • Auto-merge star history PR after creation (#280) (@houko)
  • Use PR instead of direct push for star history workflow (#279) (@houko)
  • Move Fly.io-specific badges from header to deploy form (#278) (@houko)
  • Revert wrangler-action to v3 (v4 does not exist) (#274) (@houko)
  • Add explicit Tauri version for MSI compatibility (#272) (@houko)
  • Prevent long commands from stretching deploy cards (#269) (@houko)
  • Remove unnecessary card width constraint on deploy page (#268) (@houko)
  • Consistent card widths on deploy page (#266) (@houko)
  • Allow multi-segment prerelease in semver validation (#263) (@houko)
  • Use docker run command on deploy hub (#262) (@houko)
  • Use prebuilt GHCR image in docker-compose.yml (#261) (@houko)
  • Docker deploy card links to correct README section (#260) (@houko)
  • Add catalog directory to Dockerfile (#256) (@houko)
  • Correct Railway URL and use prebuilt image for Render (#255) (@houko)
  • Deploy page home button links to deploy.librefang.ai (#254) (@houko)
  • Replace emoji with SVG icons and add home button (#253) (@houko)
  • Prevent release notes from being lost due to race condition (#252) (@houko)
  • Remove disk config for Render free tier (#247) (@houko)

Documentation

  • Use docker run across all README translations (#267) (@houko)
  • Separate Fly.io and Render deploy descriptions (#248) (@houko)

Maintenance

  • Keep machines running to avoid cold starts (#445) (@houko)
  • Auto-deploy to Fly.io on release (#429) (@houko)
  • Keep at least 1 machine running to avoid cold starts (#416) (@houko)
  • Add unit tests for channel rate limiter (#340) (@houko)
  • Add workflow_dispatch to deploy-worker (#273) (@houko)
  • Fix wrangler-action, force Node.js 24 (#271) (@houko)
  • Upgrade wrangler-action to v4 for Node.js 24 (#270) (@houko)
  • Add 'release' to allowed PR title types. (#246) (@houko)
  • Update star history workflow schedule to run hourly. (#245) (@houko)

[0.4.4] - 2026-03-15

Added

  • Add academic-researcher agent template (#391) (@houko)
  • Add code-review-checklist prompt skill (#377) (@houko)
  • Add API endpoints for managing extensions (#372) (@houko)
  • Add memory/knowledge graph export and import API (#371) (@houko)
  • Add POST/PUT/DELETE endpoints for MCP server config management (#370) (@houko)
  • Add POST/DELETE endpoints for model aliases (#364) (@houko)
  • Add GET /api/profiles/:name endpoint (#363) (@houko)
  • Add GET /api/tools/:name endpoint (#360) (@houko)
  • Add GET /api/schedules/:id endpoint (#291) (@houko)
  • Add GET /api/a2a/agents/:id endpoint (#290) (@houko)
  • Add PUT /api/cron/jobs/:id endpoint for updating cron jobs (#289) (@houko)
  • Horizontal scroll for long commands on deploy page (#276) (@houko)
  • Add tooltip for truncated commands on deploy page (#275) (@houko)
  • Add copy buttons to install commands on deploy hub (#258) (@houko)
  • Add macOS, Linux, Windows install options to deploy hub (#257) (@houko)
  • Deploy hub with multi-platform support (#251) (@houko)
  • Add GCP free-tier deployment with Terraform (#249) (@houko)
  • Support multi-bot routing per platform (#240) (@houko)

Fixed

  • Respect PORT env var for Railway/Render compatibility (#455) (@houko)
  • Stop syncing agent.toml versions with project release version (#375) (@houko)
  • Skip pre-release tags when finding previous version for changelog (#374) (@houko)
  • Catalog sync fails to parse remote files missing provider field (#362) (@houko)
  • Add reconnect logic to Matrix channel adapter (#361) (@houko)
  • Remove VOLUME directive from Dockerfile (#294) (@houko)
  • Render card empty due to nested anchor tags (#292) (@houko)
  • Log warnings instead of silently ignoring errors in API endpoints (#288) (@houko)
  • Add URL validation to A2A discover endpoint to prevent SSRF (#287) (@houko)
  • Validate environment variable names in channel config API (#286) (@houko)
  • Use agent_id path parameter in KV memory endpoints (#285) (@houko)
  • Use WEBSITE_REPO_TOKEN for star history workflow (#281) (@houko)
  • Auto-merge star history PR after creation (#280) (@houko)
  • Use PR instead of direct push for star history workflow (#279) (@houko)
  • Move Fly.io-specific badges from header to deploy form (#278) (@houko)
  • Revert wrangler-action to v3 (v4 does not exist) (#274) (@houko)
  • Add explicit Tauri version for MSI compatibility (#272) (@houko)
  • Prevent long commands from stretching deploy cards (#269) (@houko)
  • Remove unnecessary card width constraint on deploy page (#268) (@houko)
  • Consistent card widths on deploy page (#266) (@houko)
  • Allow multi-segment prerelease in semver validation (#263) (@houko)
  • Use docker run command on deploy hub (#262) (@houko)
  • Use prebuilt GHCR image in docker-compose.yml (#261) (@houko)
  • Docker deploy card links to correct README section (#260) (@houko)
  • Add catalog directory to Dockerfile (#256) (@houko)
  • Correct Railway URL and use prebuilt image for Render (#255) (@houko)
  • Deploy page home button links to deploy.librefang.ai (#254) (@houko)
  • Replace emoji with SVG icons and add home button (#253) (@houko)
  • Prevent release notes from being lost due to race condition (#252) (@houko)
  • Remove disk config for Render free tier (#247) (@houko)

Documentation

  • Use docker run across all README translations (#267) (@houko)
  • Separate Fly.io and Render deploy descriptions (#248) (@houko)

Maintenance

  • Keep machines running to avoid cold starts (#445) (@houko)
  • Auto-deploy to Fly.io on release (#429) (@houko)
  • Keep at least 1 machine running to avoid cold starts (#416) (@houko)
  • Add unit tests for channel rate limiter (#340) (@houko)
  • Add workflow_dispatch to deploy-worker (#273) (@houko)
  • Fix wrangler-action, force Node.js 24 (#271) (@houko)
  • Upgrade wrangler-action to v4 for Node.js 24 (#270) (@houko)
  • Add 'release' to allowed PR title types. (#246) (@houko)
  • Update star history workflow schedule to run hourly. (#245) (@houko)

[0.4.3-beta3] - 2026-03-14

Fixed

  • Render card empty due to nested anchor tags (#292) (@houko)
  • Use WEBSITE_REPO_TOKEN for star history workflow (#281) (@houko)
  • Auto-merge star history PR after creation (#280) (@houko)
  • Use PR instead of direct push for star history workflow (#279) (@houko)
  • Move Fly.io-specific badges from header to deploy form (#278) (@houko)

[0.4.3-beta2] - 2026-03-14

Added

  • Horizontal scroll for long commands on deploy page (#276) (@houko)
  • Add tooltip for truncated commands on deploy page (#275) (@houko)
  • Support multi-bot routing per platform (#240) (@houko)

Fixed

  • Revert wrangler-action to v3 (v4 does not exist) (#274) (@houko)
  • Add explicit Tauri version for MSI compatibility (#272) (@houko)
  • Prevent long commands from stretching deploy cards (#269) (@houko)
  • Remove unnecessary card width constraint on deploy page (#268) (@houko)
  • Consistent card widths on deploy page (#266) (@houko)
  • Use prebuilt GHCR image in docker-compose.yml (#261) (@houko)

Documentation

  • Use docker run across all README translations (#267) (@houko)

Maintenance

  • Add workflow_dispatch to deploy-worker (#273) (@houko)
  • Fix wrangler-action, force Node.js 24 (#271) (@houko)
  • Upgrade wrangler-action to v4 for Node.js 24 (#270) (@houko)

[0.4.3-beta] - 2026-03-14

Added

  • Add copy buttons to install commands on deploy hub (#258) (@houko)
  • Add macOS, Linux, Windows install options to deploy hub (#257) (@houko)
  • Deploy hub with multi-platform support (#251) (@houko)
  • Add GCP free-tier deployment with Terraform (#249) (@houko)

Fixed

  • Allow multi-segment prerelease in semver validation (#263) (@houko)
  • Use docker run command on deploy hub (#262) (@houko)
  • Docker deploy card links to correct README section (#260) (@houko)
  • Add catalog directory to Dockerfile (#256) (@houko)
  • Correct Railway URL and use prebuilt image for Render (#255) (@houko)
  • Deploy page home button links to deploy.librefang.ai (#254) (@houko)
  • Replace emoji with SVG icons and add home button (#253) (@houko)
  • Prevent release notes from being lost due to race condition (#252) (@houko)
  • Remove disk config for Render free tier (#247) (@houko)

Documentation

  • Separate Fly.io and Render deploy descriptions (#248) (@houko)

Maintenance

  • Add 'release' to allowed PR title types. (#246) (@houko)
  • Update star history workflow schedule to run hourly. (#245) (@houko)

[0.4.2] - 2026-03-14

Added

  • Add CLI deploy command and FAQ to deploy page (#238) (@houko)
  • Auto-sync model catalog on daemon startup (#237) (@houko)
  • Add channel sidecar protocol for external adapters (#228) (@houko)
  • Integrate model-catalog sync with dashboard UI (#227) (@houko)
  • Add cargo feature flags for channel adapters (#223) (@houko)
  • Improve community organization and version governance (#212) (@houko)

Fixed

  • Revert file versions to 0.4.1-20260314 and fix release.sh (#243) (@houko)
  • Release script uses PR instead of direct push (#242) (@houko)
  • Daemon env vars, MCP probe, and SSE parsing (#211) (@houko)

Changed

  • Replace hardcoded model catalog with include_str TOML (#235) (@houko)
  • Replace provider match with static registry (#224) (@houko)

Documentation

  • Add integration test writing guide to CONTRIBUTING.md (#232) (@houko)
  • Add channel adapter contribution example (#231) (@houko)

Maintenance

  • Bump version to v0.4.2-20260314 (#241) (@houko)
  • Trigger deploy worker auto-deploy (#239) (@houko)
  • Add pre-commit hooks and i18n contribution guide (#233) (@houko)
  • Add justfile for unified dev commands (#230) (@houko)
  • Upgrade GitHub Actions for Node.js 24 compatibility (#229) (@houko)

[0.4.0] - 2026-03-14

Authentication & Drivers

  • ChatGPT Session Auth: New browser-based OAuth flow for ChatGPT Plus/Ultra subscribers.
    • PKCE S256 code challenge for secure token exchange.
    • Automatic model discovery (Codex endpoints).
    • librefang auth chatgpt subcommand to easily link accounts.
    • Persistent session caching with 7-day TTL.
  • MiniMax Dual-Platform Support: Added separate minimax-cn provider for China-specific endpoints (using MINIMAX_CN_API_KEY).
  • QQ Bot Adapter: Native support for QQ Bot messaging channel.

Web Dashboard & i18n

  • Internationalization (i18n): Full support for multiple languages in the dashboard.
    • Added zh-CN (Simplified Chinese) locale.
    • Unified translation helper t() across all JS modules.
  • UI Overhaul:
    • New sidebar layout with integrated theme/language switchers.
    • Replaced emoji icons with high-quality inline SVG icons (globe, search, chart, etc.).
    • Improved ClawHub category wrapping for better responsiveness on small screens.

Core Platform

  • Version Alignment: Synced all 31 built-in agents and sub-packages to version 0.4.0.
  • Config Hot-Reloading: Enhanced reliability for runtime configuration updates without daemon restarts.

0.1.0 - 2026-02-24

Core Platform

  • 15-crate Rust workspace: types, memory, runtime, kernel, api, channels, wire, cli, migrate, skills, hands, extensions, desktop, xtask
  • Agent lifecycle management: spawn, list, kill, clone, mode switching (Full/Assist/Observe)
  • SQLite-backed memory substrate with structured KV, semantic recall, vector embeddings
  • 41 built-in tools (filesystem, web, shell, browser, scheduling, collaboration, image analysis, inter-agent, TTS, media)
  • WASM sandbox with dual metering (fuel + epoch interruption with watchdog thread)
  • Workflow engine with pipelines, fan-out parallelism, conditional steps, loops, and variable expansion
  • Visual workflow builder with drag-and-drop node graph, 7 node types, and TOML export
  • Trigger system with event pattern matching, content filters, and fire limits
  • Event bus with publish/subscribe and correlation IDs
  • 7 Hands packages for autonomous agent actions

LLM Support

  • 3 native LLM drivers: Anthropic, Google Gemini, OpenAI-compatible
  • 27 providers: Anthropic, Gemini, OpenAI, Groq, OpenRouter, DeepSeek, Together, Mistral, Fireworks, Cohere, Perplexity, xAI, AI21, Cerebras, SambaNova, Hugging Face, Replicate, Ollama, vLLM, LM Studio, and more
  • Model catalog with 130+ built-in models, 23 aliases, tier classification
  • Intelligent model routing with task complexity scoring
  • Fallback driver for automatic failover between providers
  • Cost estimation and metering engine with per-model pricing
  • Streaming support (SSE) across all drivers

Token Management & Context

  • Token-aware session compaction (chars/4 heuristic, triggers at 70% context capacity)
  • In-loop emergency trimming at 70%/90% thresholds with summary injection
  • Tool profile filtering (cuts default 41 tools to 4-10 for chat agents, saving 15-20K tokens)
  • Context budget allocation for system prompt, tools, history, and response
  • MAX_TOOL_RESULT_CHARS reduced from 50K to 15K to prevent tool result bloat
  • Default token quota raised from 100K to 1M per hour

Security

  • Capability-based access control with privilege escalation prevention
  • Path traversal protection in all file tools
  • SSRF protection blocking private IPs and cloud metadata endpoints
  • Ed25519 signed agent manifests
  • Merkle hash chain audit trail with tamper detection
  • Information flow taint tracking
  • HMAC-SHA256 mutual authentication for peer wire protocol
  • API key authentication with Bearer token
  • GCRA rate limiter with cost-aware token buckets
  • Security headers middleware (CSP, X-Frame-Options, HSTS)
  • Secret zeroization on all API key fields
  • Subprocess environment isolation
  • Health endpoint redaction (public minimal, auth full)
  • Loop guard with SHA256-based detection and circuit breaker thresholds
  • Session repair (validates and fixes orphaned tool results, empty messages)

Channels

  • 40 channel adapters: Telegram, Discord, Slack, WhatsApp, Signal, Matrix, Email, Teams, Mattermost, Google Chat, Webex, Feishu/Lark, LINE, Viber, Facebook Messenger, Mastodon, Bluesky, Reddit, LinkedIn, Twitch, IRC, XMPP, and 18 more
  • Unified bridge with agent routing, command handling, message splitting
  • Per-channel user filtering and RBAC enforcement
  • Graceful shutdown, exponential backoff, secret zeroization on all adapters

API

  • 100+ REST/WS/SSE API endpoints (axum 0.8)
  • WebSocket real-time streaming with per-agent connections
  • OpenAI-compatible /v1/chat/completions API (streaming SSE + non-streaming)
  • OpenAI-compatible /v1/models endpoint
  • WebChat embedded UI with Alpine.js
  • Google A2A protocol support (agent card, task send/get/cancel)
  • Prometheus text-format /api/metrics endpoint for monitoring
  • Multi-session management: list, create, switch, label sessions per agent
  • Usage analytics: summary, by-model, daily breakdown
  • Config hot-reload via polling (30-second interval, no restart required)

Web UI

  • Chat message search with Ctrl+F, real-time filtering, text highlighting
  • Voice input with hold-to-record mic button (WebM/Opus codec)
  • TTS audio playback inline in tool cards
  • Browser screenshot rendering in chat (inline images)
  • Canvas rendering with iframe sandbox and CSP support
  • Session switcher dropdown in chat header
  • 6-step first-run setup wizard with provider API key help (12 providers)
  • Skill marketplace with 4 tabs (Installed, ClawHub, MCP Servers, Quick Start)
  • Copy-to-clipboard on messages, message timestamps
  • Visual workflow builder with drag-and-drop canvas

Client SDKs

  • JavaScript SDK (@librefang/sdk): full REST API client with streaming, TypeScript declarations
  • Python client SDK (librefang_client): zero-dependency stdlib client with SSE streaming
  • Python agent SDK (librefang_sdk): decorator-based framework for writing Python agents
  • Usage examples for both languages (basic + streaming)

CLI

  • 14+ subcommands: init, start, agent, workflow, trigger, migrate, skill, channel, config, chat, status, doctor, dashboard, mcp
  • Daemon auto-detection via PID file
  • Shell completion generation (bash, zsh, fish, PowerShell)
  • MCP server mode for IDE integration

Skills Ecosystem

  • 60 bundled skills across 14 categories
  • Skill registry with TOML manifests
  • 4 runtimes: Python, Node.js, WASM, PromptOnly
  • FangHub marketplace with search/install
  • ClawHub client for OpenClaw skill compatibility
  • SKILL.md parser with auto-conversion
  • SHA256 checksum verification
  • Prompt injection scanning on skill content

Desktop App

  • Tauri 2.0 native desktop app
  • System tray with status and quick actions
  • Single-instance enforcement
  • Hide-to-tray on close
  • Updated CSP for media, frame, and blob sources

Session Management

  • LLM-based session compaction with token-aware triggers
  • Multi-session per agent with named labels
  • Session switching via API and UI
  • Cross-channel canonical sessions
  • Extended chat commands: /new, /compact, /model, /stop, /usage, /think

Image Support

  • ContentBlock::Image with base64 inline data
  • Media type validation (png, jpeg, gif, webp only)
  • 5MB size limit enforcement
  • Mapped to all 3 native LLM drivers

Usage Tracking

  • Per-response cost estimation with model-aware pricing
  • Usage footer in WebSocket responses and WebChat UI
  • Usage events persisted to SQLite
  • Quota enforcement with hourly windows

Interoperability

  • OpenClaw migration engine (YAML/JSON5 to TOML)
  • MCP client (JSON-RPC 2.0 over stdio/SSE, tool namespacing)
  • MCP server (exposes LibreFang tools via MCP protocol)
  • A2A protocol client and server
  • Tool name compatibility mappings (21 OpenClaw tool names)

Infrastructure

  • Multi-stage Dockerfile (debian:bookworm-slim runtime)
  • docker-compose.yml with volume persistence
  • GitHub Actions CI (check, test, clippy, format)
  • GitHub Actions release (multi-platform, GHCR push, SHA256 checksums)
  • Cross-platform install script (curl/irm one-liner)
  • systemd service file for Linux deployment

Multi-User

  • RBAC with Owner/Admin/User/Viewer roles
  • Channel identity resolution
  • Per-user authorization checks
  • Device pairing and approval system

Production Readiness

  • 1731+ tests across 15 crates, 0 failures
  • Cross-platform support (Linux, macOS, Windows)
  • Graceful shutdown with signal handling (SIGINT/SIGTERM on Unix, Ctrl+C on Windows)
  • Daemon PID file with stale process detection
  • Release profile with LTO, single codegen unit, symbol stripping
  • Prometheus metrics for monitoring
  • Config hot-reload without restart