What's Changed

First stable release of the v1.5.0 series since v1.4.1. Stabilizes v1.5.0-beta.1 → beta.5 (~93 commits) with production-validated P1 stability fixes (#1693 fixing #1686).

If you have been running any 1.5.0-beta.* in production, upgrading to v1.5.0 is recommended — same runtime as beta.5 with stable versioning and npm @latest.

New Features

• New platforms — Tencent Yuanbao Bot API (#1445); cloud_web self-hosted IM gateway (#1282); Google Chat (#1424); WPS Agentspace (#1439); Tuitui (#849)
• New agent — Reasonix HTTP serve API adapter (#1281)
• Feishu hardening — topic workspace isolation (#1551); quoted file on-demand download (#1588); first thread mention bootstrap (#1627); cross-type image batch flush (#1693)
• Agent ecosystem — Kimi Code CLI native dialect (#1564); Pi v0.84.0 toolcall_end + willRetry (#1674, #1597); codex /model gpt-5.x visibility (#1546); Claude Code session title fallback (#1549)
• Pi & permissions — RPC mode (#1440); permission mode env injection (#1637); agent_session_idle_timeout_mins (#1338)

Fixed

• #1686 P1-A — /restart no longer crashes daemon when platform adapter panics during post-restart notification (#1693)
• #1686 P1-B — Rapid image + text within image_batch_window_ms no longer drops the buffered image (#1693)
• #1686 P1-C — Long background turns no longer killed by stale idle-close timer (#1693)
• Weixin send budget — sendmessage ret=-2 fails fast per-account (#1643)
• Core stability — idle close race fixes; attachment filename collision prevention; code fence split integrity (#1630, #1557)

Breaking Changes

None. All new config options are optional with safe defaults. v1.4.1 configs upgrade as-is.

Full Changelog: v1.4.1...v1.5.0

v1.5.0 Latest

Latest

Compare

Choose a tag to compare

Rolling beta with 1 commit on top of v1.5.0-beta.4: P1 stability fixes (#1693 fixing #1686).

Fixed

• Restart-notify panic recovery — daemon no longer crashes if platform panics during post-restart notification
• Cross-type image batch flush — prevents image loss when text arrives during image_batch_window_ms
• Idle close race fixes — background turns re-arm idle timer; schedule before drain

See changelogs/v1.5.0-beta.5.md for details.

Full Changelog: v1.5.0-beta.4...v1.5.0-beta.5

v1.5.0-beta.5 Pre-release

Pre-release

Compare

Choose a tag to compare

Rolling beta with 6 commits from main since v1.5.0-beta.3 (2026-08-14).

Features

• Kimi Code CLI native dialect (#1564) — full support for Node.js kimi-code: probe-gated flags, modern resume, dual stream parser, session listing

Fixed

• Kimi --work-dir flag gating (#1483) — fixes unknown option --work-dir on modern CLI
• Pi willRetry turn continuity (#1597) — keep turn open during auto-retry
• Pi v0.84.0 toolcall_end (#1674) — handle events without cumulative message
• WeCom quoted message context (#1669)
• Web admin work_dir validation (#1572)

See changelogs/v1.5.0-beta.4.md for details.

Full Changelog: v1.5.0-beta.3...v1.5.0-beta.4

v1.5.0-beta.4 Pre-release

Pre-release

Compare

Choose a tag to compare

Rolling beta with 36 commits merged from main since v1.5.0-beta.2 (2026-07-14). Owner decision: ship beta.3 before stable — beta.2 is ~1 month old.

New Platforms

• Google Chat — first-class platform adapter (#1424)
• WPS Agentspace — auto-login and token encryption (#1439)
• Tuitui — platform support (#849)

Features

• Feishu topic workspace isolation (#1551)
• Feishu quoted file on-demand download (#1588)
• Feishu first thread mention bootstrap (#1627)
• DingTalk chat-list title from content (#1288)
• Antigravity tool permissions bridge (#1328)
• Admin-gated exec commands — /commands addexec and /cron addexec behind admin_from (#1036)
• Pi permission mode env injection (#1637)
• Agent cmd TOML array form (#1673)
• sonnet[1m] fallback model (#1107)

Fixed

• Weixin send budget ret=-2 — fail fast per-account (#1643)
• Web admin agent-type-aware permission modes (#1251)
• Pi models-store.json fallback (#1636)
• Core code fence split (#1630)
• Core tool_max_len in card mode (#1257)
• DingTalk @userid extraction (#1250)
• Attachment filename collisions (#1557)
• Claude Code session title fallback (#1549)
• Claude Code missing work_dir detection (#1425)
• Antigravity session resume (#1584)
• Usage 7-day window duplicate render (#1583)
• i18n /model switch copy (#1373)
• CLI reject unknown commands (#353)
• Kimi preset global/China split
• cloud-web IPv6 test fix (#1646)
• darwin cross-build — remove duplicate CheckLinger stub that broke macOS release binary (#1093 follow-up)

See changelogs/v1.5.0-beta.3.md for full details.

Full Changelog: v1.5.0-beta.2...v1.5.0-beta.3

v1.5.0-beta.3 Pre-release

Pre-release

Compare

Choose a tag to compare

Rolling beta with 1 commit on top of v1.5.0-beta.1: codex /model gpt-5.x visibility fix.

Fixed

• codex /model gpt-5.x visibility (#1546 / cherry-pick #1547, @chenhg5)
  • Replaces stale 11-entry static allowlist (unchanged since 2026-03) with prefix pattern matching
  • Users on codex-cli ≥ 0.143 could not select GPT-5.6 (gpt-5.6-sol/terra/luna) in /model
  • Only affects API-fetch fallback path; model_catalog_json / explicit models = [...] configs unaffected
  • Owner E2E validated on Feishu test-feishu-codex; 45 new unit tests

Install

npm install -g cc-connect@beta
# or
npm install -g [email protected]

Full changelog: changelogs/v1.5.0-beta.2.md
Diff: v1.5.0-beta.1...v1.5.0-beta.2

v1.5.0-beta.2 Pre-release

Pre-release

Compare

Choose a tag to compare

First beta of the v1.5.0 series since v1.4.1 stable. Three new integration surfaces join the family.

Highlights

New Platforms

• Tencent Yuanbao Bot API — first-class Yuanbao platform adapter (#1445, @skyblue)
• cloud_web — self-hosted IM gateway (CWIP v1; websocket / long_poll / gateway). See docs/cloud-web.md (#1282, @jiagou123)

New Agent

• Reasonix — HTTP serve API adapter with default/yolo/plan permission modes (#1281, @mchenziyi)

Features

• agent_session_idle_timeout_mins — close idle live agent processes after a clean turn while preserving session (#1338, @hl1221hl)
• Pi RPC mode with extension_ui permission forwarding (#1440, @happyTonakai)
• Feishu mention_map + inter-bot relay (#1341, #1413)
• Russian (ru) i18n for Web admin UI (#1449, @sonsay)

Fixes

• Resume stream preview after permission prompt (#1451)
• Absolute paths from SaveFilesToDisk / AppendFileRefs (#1462)
• Queue placeholder before session lock (#1389)
• Codex app-server write timeout (#1448)
• Slack streaming-card NO_REPLY suppress (#1397)
• Hide agent footer lines (#1416)
• Release branch: Windows build fix for Pi unused import (#pi proc_windows)

Install

npm install -g cc-connect@beta
# or
npm install -g [email protected]

Full changelog: changelogs/v1.5.0-beta.1.md

v1.5.0-beta.1 Pre-release

Pre-release

Compare

Choose a tag to compare

Patch release focused on Kimi CLI --print compatibility for users on the newer kimi-code 0.14.x (which removed the --print flag).

This is the first release under the post-v1.4.0 SOP correction: every commit since v1.4.0 (here #1461) went through a fresh manual QA cycle by cc-connect/qa-cursor before stable promotion, including owner-paired binary smoke against both old and new Kimi CLI versions.

🔧 Fixed

• Kimi CLI --print compatibility (#1461 fixing #1456, @chenhg5, reported by @WeiFengJL)
  • Newer kimi-code 0.14.x removed the --print flag, causing error: unknown option '--print' (Did you mean --prompt?) on agent startup.
  • cc-connect now probes Kimi CLI --help at startup and conditionally passes --print only when the installed binary supports it.
  • Both legacy kimi-cli 1.48.x users and new kimi-code 0.14.x users upgrade to v1.4.1 with zero config changes.
  • Verified against real binaries: kimi-code 0.14.2 (Print=false, unblocked) and kimi-cli 1.48.0 (Print=true, no regression).

Upgrade notes

• No config changes required. Drop in v1.4.1 and Kimi CLI works on both old and new versions.
• All v1.4.0 upgrade notes still apply (cli_path deprecation, Feishu imageBatchWindow / MessageRecallDetector defaults).

Install

# npm
npm install -g cc-connect

# Binary: download from assets below

Full changelog: changelogs/v1.4.1.md
Diff from v1.4.0: v1.4.0...v1.4.1

v1.4.1

Compare

Choose a tag to compare

Stable release of the v1.4.0 series. Two new platforms join the family (Cisco Webex, Matrix with E2EE), broader configurability across agents and platforms, Korean i18n, and a long list of fixes including three critical post-beta cherry-picks.

🚨 Critical fixes shipped late

• core Send-goroutine nil-pointer panic race (#1436, @gotang) — would crash the whole cc-connect process and drop every platform connection. 100% reproducible in production.
• Feishu recall-probe quota burn (#1321, @qvictl) — MessageRecallDetector was burning ~1.3M Feishu OpenAPI calls/month per stuck session. Probe interval now 1 min with dedup.
• claudecode run_as_user EACCES regression (#1433, @chenhg5; reported by @vuyiv #1429) — run_as_user users on beta.1/beta.2 were 100% blocked at agent startup. Fix: chmod 0o644 on per-spawn system-prompt temp file.

Highlights

• New platforms: Cisco Webex first-class adapter (#1402, @bryantbarzola); Matrix with E2EE (#834). 15 platforms total.
• agent option parsing refactor — unified cmd field across all agent adapters; cli_path deprecated (#1297, @happyTonakai).
• Slack streaming preview + aggregated turn card (#1333).
• Feishu after_click card replacement for cmd: actions (#1299); batch images coalesced (default 500 ms window) (#1408 carrying #1395).
• Codex system_prompt / append_system_prompt (#1345); Codex model_catalog_json highest-priority (#1074).
• Zhipu GLM provider presets (z.ai + bigmodel CN) (#1412, @clingnet).
• Korean (ko) i18n for Web admin UI (#1343).
• Claude Code plugin_dir (#1325), cc-connect send --cwd (#1380, @MMMarcinho), max_attachment_size_mb (#1392, @rablwupei), CC_LOG_MAX_BACKUPS env (#1260), configurable /history truncation (#1291, @AaronZ345).
• 30+ fixes across feishu, slack, dingtalk, claudecode, codex, core engine, runas, web admin and i18n.

Upgrade notes

• cli_path deprecated → switch to unified cmd (existing configs keep working with a deprecation warning).
• Feishu imageBatchWindow default: 150 ms → 500 ms.
• Feishu MessageRecallDetector fallback probe interval: 2 s → 60 s.

Install

# npm
npm install -g cc-connect

# Binary: download from assets below

Full themed changelog: changelogs/v1.4.0.md
Full diff: v1.3.4...v1.4.0

Thanks to all 11 external contributors and the cc-connect team for this release. 🎉

v1.4.0

Compare

Choose a tag to compare

Rolling beta with 3 commits on top of v1.4.0-beta.2. Includes one critical
run_as_user regression fix.

🚨 Critical Fix

• claudecode run_as_user EACCES regression — chmod 0o644 on per-spawn
  system-prompt temp file so non-root child processes can read it after dropping
  privileges. Regression introduced by v1.3.4 (#1376) when Windows cmdline hotfix
  switched to file-based prompt with default 0o600 perms.
  Impact: 100% of run_as_user users blocked at agent startup on beta.1/beta.2.
  Reported by @vuyiv (#1429), fixed by @chenhg5 (#1433).

Features

• Zhipu GLM provider presets — add z.ai and bigmodel (CN) preset entries
  to embedded provider-presets.json. Pure additive (#1412, @clingnet).

Chore

• .gitignore: add .worktrees/ for local-only multi-agent scratch dirs (#1443).

Install

# npm (beta channel)
npm install -g cc-connect@beta

# Binary: download from assets below

Full changelog: changelogs/v1.4.0-beta.3.md
Diff from beta.2: v1.4.0-beta.2...v1.4.0-beta.3

v1.4.0-beta.3 Pre-release

Pre-release

Compare

Choose a tag to compare

Rolling beta with 10 additional commits on top of v1.4.0-beta.1:
3 QA-found hotfixes + 7 cherry-picked PRs from main. All changes verified by QA.

Highlights

Features

• codex model_catalog_json — prefer Codex's own model_catalog_json as highest-priority model source (#1074)
• cc-connect send --cwd — specify the working directory for send commands (#1380)

Refactor

• agent option parsing — centralize cmd/env option parsing into core with unified cmd field (+863/-419 across 39 files). All agent startup paths now share the same option parsing logic (#1297) ⚠️ Monitor agent startup compatibility with existing configs.

Bug Fixes

• feishu image batch window — configurable; default 150→500ms (QA hotfix)
• slack streaming card — stop chat.update on payload size limit; fallback to postMessage (QA hotfix)
• i18n nav.cron — translate for ko/ja/es (QA hotfix)
• /history truncation — configurable length (#1291)
• post-restart notification — queue and dispatch on platform ready (#1388)
• claudecode progress card — emit EventToolResult (#1407)
• dingtalk stream panic — recover panic in processLoop (#1390)

Install

# npm (beta channel)
npm install -g cc-connect@beta

# Binary: download from assets below

Full changelog: changelogs/v1.4.0-beta.2.md
Diff from beta.1: v1.4.0-beta.1...v1.4.0-beta.2

v1.4.0-beta.2 Pre-release

Pre-release

Compare

Choose a tag to compare

First stable release of the 1.3.3 series. Stabilizes the v1.3.3-beta.1 → v1.3.3-beta.5 line (≈ 235 PRs since v1.3.2) plus 7 post-beta.5 fixes. See changelogs/v1.3.3.md for the full themed summary; per-beta details remain in the beta sections below.

Highlights

• New agents: Devin CLI, Google Antigravity (agy), GitHub Copilot — all first class. Hardened coverage for Cursor, OpenCode, Qoder, Kimi, Pi.
• Platform expansion: QQ Bot inline keyboards + file send/receive (OneBot), WeCom SendFile (WebSocket), Feishu audio + video native media, Slack Assistant API, MAX webhook delivery, DingTalk @mentions / richText / image / file inbound, WPS Xiezuo (金山协作), broader Weibo DM.
• Long-running turn hardening: new max_turn_time_mins wall-clock cap with soft-stop + force-kill + auto-resume — long bash / test commands can no longer lock a session indefinitely.
• Core commands: /timer, /cancel, /ps (replaces /btw), cron add --silent, agent-driven TTS.
• Multi-user / permissions: reply-to-unauthorized-IM-senders, @mention-tolerant permission keywords (@Bot/permit ≡ /permit), Bridge requires token when enabled.
• Observability: blackbox testing framework (P0/P1/P2 + config-switch matrix), CUJ test framework, agent-resume regression suite, Pi context-usage reporter.
• Provider ecosystem: NekoCode, VisionCoder, AIHubMix, MiniMax M3 presets; Claude Code 1M-context Opus + append_system_prompt + PermissionRequest hooks; Codex request_user_input app-server events; configurable shell + shell profile.

Post-beta.5 fixes (delta from beta.5)

• qoder: emit streaming text without dropping final result (#1290)
• weixin: use ilink_user_id in getConfigReq for typing ticket (#1308)
• daemon: remove redundant linger_other.go that breaks non-linux builds (#1314)
• wps-xiezuo: preserve newlines in outbound messages — fixes unreadable /status (#1361)
• core: /switch no longer loses history; persist user msgs immediately; add CUJ test framework (#1348)
• core: queued FIFO drains no longer drop earlier queued messages as stale just because a later queued message has a higher create_time (#1286)
• core: make /history entry truncation configurable via [display].history_max_len, defaulting to 1000; 0 disables truncation (#1291)
• tts/minimax: drop status=2 trailer chunk to stop audio playing twice (#1364)
• tests: add provider-resume regression tests for codex / opencode / kimi (#1366)

⚠️ Behavior Changes (carried forward from the beta cycle)

All behavior changes from beta.1 → beta.5 remain in effect for v1.3.3. Most likely to affect existing configs:

• progress_style default for Telegram & Discord is now compact (was legacy). Set progress_style = "legacy" to revert. (#1354)
• QQ Bot default intents now include INTERACTION_CREATE (bit 26). Custom intents must include 1<<26 for inline keyboard buttons.
• DingTalk msgtype=file inbound now reaches the agent (#1357).
• Engine permission keyword matching is @mention-tolerant: @Bot/permit ≡ /permit (#1358).
• reset_on_idle_mins default is now 30 minutes (#494). Set to 0 to disable.
• Bridge with no [bridge].token configured will refuse to start (#408).

Breaking Changes

None. Fully additive release.

Upgrade

npm i -g [email protected]
# or
go install github.com/chenhg5/cc-connect/cmd/[email protected]

Coming from a v1.3.3-beta.*: this is a small fix-only upgrade. No config change required.

Coming from v1.3.2: review the Behavior Changes above before upgrading.

Large beta with 74 PRs from 28 contributors. New agents (Google Antigravity agy, GitHub Copilot), QQ file send/receive via OneBot, WeCom SendFile (WebSocket), Feishu audio/video media, agent-driven TTS, /timer and /cancel commands, and broad platform fixes across Telegram, Discord, DingTalk, Feishu, WeCom, WeiXin, Cursor, OpenCode, Pi, and Codex. See changelogs/v1.3.3-beta.5.md for the full list.

New Features

• Google Antigravity (agy) agent as a first-class integration (#1123)
• GitHub Copilot agent as a first-class integration (#865)
• /timer — one-shot delayed task system (#1012)
• /cancel — interrupt and reset the current session (#957)
• Session prune command to remove duplicate sessions (#603)
• Agent-driven TTS send (#1230)
• Reply to unauthorized IM senders option (#1190)
• QQ Bot inline keyboard buttons and INTERACTION_CREATE events. Permission requests now render as clickable buttons (#1131)
• QQ (OneBot) file send/receive via HTTP API (#323)
• WeCom SendFile in WebSocket mode (#1199)
• Feishu audio + video attachments as native media (#1202)
• Feishu rich card rendering + panel handling refresh (#1204)
• DingTalk reaction emoji support (#1213)
• DingTalk @mention via send --at-users / --at-all (#1188)
• Slack + tmux per-thread session scope with per-session tmux windows (#1179)
• cron add --silent CLI flag (#1285, closes #858)
• Codex request_user_input app-server events + relay group visibility (#1200, #1209)
• Claude Code custom append_system_prompt + PermissionRequest hooks (#1175, #850)
• Pi ContextUsageReporter for reply footer token stats (#1235)
• Daemon hardened service-file env capture + EnvDiscoverer plugin hook (#1034)
• Configurable shell + shell profile for exec (#870)

Fixed

• Many fixes across the engine, agents, and platforms. Highlights: Telegram/Discord progress style, DingTalk file inbound, Feishu link/card URLs, WeCom long-message split, Cursor session titles, OpenCode tool rejection, Codex resume + sandbox_mode, Pi /dir and /model, Windows instance lock, and Claude Code provider preservation. See changelogs/v1.3.3-beta.5.md.

⚠️ Behavior Notes

• progress_style default for Telegram and Discord is now compact (was legacy). Set progress_style = "legacy" in the platform config to restore previous behavior (#1354).
• DingTalk msgtype=file inbound messages now reach the agent. Previously silently dropped (#1357).
• Engine permission keyword matching is now @mention-tolerant: @Bot/permit matches the same as /permit (#1358).

⚠️ QQ Bot Intent Configuration

The default intents for QQ Bot now include INTERACTION_CREATE (bit 26, value 1<<26). If you previously set a custom intents value without this bit, inline keyboard buttons will not work — update your intents to include bit 26. If you use the default intents, no action is needed. See config.example.toml for the new intents option.

New Features

• max_turn_time_mins: new config option — absolute wall-clock cap per agent turn that does NOT reset on tool-call events. Prevents long-running bash commands from permanently locking the session (#1091). Uses a two-phase shutdown: soft stop (10s grace) then force-kill. Session is preserved and resumed via --resume on the next message.

Fixed

• Web console 404 regression: make release-all did not depend on make web, so release binaries were built without frontend assets when web/dist/ was empty (gitignored). All routes on the management port returned 404. Fixed by adding web as a prerequisite of release-all (#1136)
• Slack @mention without space: stripAppMentionText only matched "> " (with trailing space), so @Bot/command (no space) was forwarded verbatim to Claude instead of being parsed as a command
• DingTalk msgtype="picture" dropped: image messages delivered as "picture" (instead of "image") were silently dropped. Both types now route to the image handler (#1128)
• Feishu require_mention = false ignored: the platform read group_reply_all but users set require_mention = false; now both are treated as equivalent (#1141)
• AskUserQuestion resolved with empty answer: delivery receipts and read-notifications (empty messages) were accepted as valid answers to AskUserQuestion, resolving it within ~500ms before the user could respond. Empty/whitespace content is now rejected (#1086)

Beta release with blackbox testing infrastructure, cursor/opencode agent support, and bug fixes.

New Features

• Blackbox testing framework: Phase 1-2 blackbox testing with P0/P1/P2 coverage, config-switch, and NewEnvWithSetup infrastructure
• Cursor/OpenCode agents: add cursor and opencode agent support in blackbox tests

Fixed

• Core italic wrapping: restore italic wrapping on reply footer
• Feishu footer asterisks: strip asterisks from footer to prevent Feishu markdown italic rendering
• Kimi session UUID: capture session UUID from stderr instead of stdout
• Codex stdio sentinel: add stdio sentinel for Codex app_server backend
• Windows cross-compile: add missing CheckLinger stub to daemon/windows.go and daemon/unsupported.go so make release-all succeeds for all target platforms

Beta release with Slack Assistant API, DingTalk improvements, MAX platform webhook mode, and numerous platform fixes. No breaking changes.

New Features

• Slack Assistant API: support Slack Assistant API (Agent toggle) with natural on/off switching (#844)
• DingTalk richText: support richText message type for DingTalk platform (#828)
• DingTalk image handling: add DingTalk image message support (#828)
• MAX webhook delivery mode: add webhook delivery mode for MAX messenger platform with deployment docs (#818)
• Claude Code env vars: support project-level environment variables via env config section (#812)
• display_mode enum: add display_mode enum to replace boolean quiet config, with quiet/compact/normal/full options (#655)
• Core reset_on_idle_mins default: default to 30 minutes to prevent context drift (#494)
• Claude Code custom system prompt: add support for custom system prompt configuration via system_prompt option (#534)

Fixed

• Bridge security: require token when Bridge is enabled to prevent unauthorized access (#408)
• Feishu recalled messages: handle recalled messages gracefully (#841)
• Feishu media download failure: notify user when media download fails instead of silent drop (#815)
• WeChat video messages: send video files as proper video messages in WeChat (#813)
• WeChat incomplete delivery: notify user on incomplete message delivery and enhance retry logging (#771)
• Telegram private topics: preserve private topic session keys (#804)
• Kimi session UUID: capture session UUID from stderr instead of stdout (#766)
• Codex app_server config: app_server backend should honor model/effort/provider config + add stdio sentinel (#837)
• Codex progress rendering: render progress in rich Card 2.0 format (#838)
• Core ellipsis events: suppress ellipsis-only events and handle context indicator in footer
• Core Markdown table: render inline formatting inside GFM table cells (#675)
• Feishu user id resolution: guard user id resolution against edge cases
• Feishu thread topics: skip quote injection in thread-isolated topics (#767)
• Config display mode: honor project display mode setting
• Daemon restart: add --force flag to daemon restart command (#736)
• AskUserQuestion: use question text as answers key for proper answer routing (#822)

Beta release with new agents, new features, and broad platform fixes. No breaking changes.

New Features

• Devin agent: add Devin CLI as a first-class agent with full /list, /mode, and session management (#672)
• /ps command (replaces /btw): send a message to a busy session mid-turn; /btw kept as alias for backward compatibility (#620)
• ! shell shortcut: use !ls -la as shorthand for /shell ls -la, with optional --timeout parameter (#658)
• NO_REPLY suppression: agents can return NO_REPLY to silently skip platform delivery, useful for cron/analysis tasks (#682)
• Feishu shared WebSocket: multiple projects sharing the same app_id now share one WebSocket connection with per-project allow_chat / group_only filtering (#613)
• Message queue depth configurable: new [queue] max_depth config option (default 5) (#690)
• Claude Code opus[1m]: add 1M-context Opus model option with shorthand descriptions (#660)
• QQ Bot file send/receive: full file attachment support with robustness checks (#685)
• Bridge ImageSender/FileSender: cc-connect send --image/--file now works through bridge protocol (#712)
• Provider presets: add NekoCode, VisionCoder, and AIHubMix to provider presets; add Trae CLI ACP and COCO ACP config examples (#739)

Fixed

• OpenCode image handling: inbound images from WeChat/WeCom are now correctly passed to OpenCode CLI via --file flags (#717)
• Slack Markdown: convert standard Markdown to Slack mrkdwn format (bold, italic, strike, links, headings) (#680)
• QQ Bot reconnect: cancel stale goroutines on WebSocket reconnect to prevent race conditions (#678)
• Gemini multiline prompt: pass prompt via stdin to preserve newlines (#695)
• Telegram HTML fallback: upgrade silent HTML parse failures to Warn-level logs (#674)
• Telegram /skills: show Telegram-safe skill command format (#571)
• Feishu webhook mode: skip bot open_id fetch in webhook mode for private deployments (#696)
• Reply footer: suppress footer when only workdir is known (#701)
• Web UI add-platform: fix "project not found" error when adding a new platform to an uncreated project

Contributors

Thanks to all contributors who made this release possible:

• @YoungShook — Devin agent integration, Telegram HTML fallback
• @Cigarrr — /ps command, NO_REPLY feature
• @vinnyxiong — Feishu shared WebSocket and allow_chat
• @happyTonakai — Shell ! prefix and --timeout
• @AaronZ345 — Claude Code opus[1m] model
• @ferocknew — QQ Bot file support
• @soaringk — OpenCode image fix
• @Zx55 — Telegram /skills fix
• @zhaomoran — Feishu webhook mode fix
• @LyInfi — Reply footer suppression
• @meloalright — Trae/COCO ACP config examples

Hotfix release: session filtering is now configurable and defaults to showing all sessions.

Fixed

• /list shows all sessions by default: the session filter introduced in v1.3.0 (which hid sessions not created by cc-connect) was accidentally merged and caused confusion. The filter is now off by default — /list, /switch, and /delete show all agent sessions regardless of origin.

Added

• filter_external_sessions config option: users who do want to hide externally-created sessions can set filter_external_sessions = true in [[projects]] to restore the old filtering behavior.
• Comprehensive integration tests: real-agent E2E tests for both Codex and Claude Code covering the full /list → /new → conversation → /list lifecycle with provider-based authentication (no env-var API keys required). Plus 9 adapter-level filter tests using real Codex/Claude Code session file fixtures.

Patch release with critical bug fixes for session management, config preservation, and Weibo media support.

Fixed

• Session visibility (/list): historical Codex sessions disappeared after upgrade due to AgentSessionID being cleared on /new or provider switch without preservation. Added PastAgentSessionIDs tracking with legacy data migration so existing sessions remain visible.
• Session naming (/new xxx): custom session names from /new were not mapped to the agent session ID for agents where the ID is established asynchronously (Codex, Qoder, Kimi, etc.). Added name mapping to all EventResult and EventText handlers across interactive, relay, and drain paths.
• Config comment preservation: /provider switch, /model, /lang, display settings, and TTS changes now use surgical text-level editing instead of full TOML re-serialization, preserving all comments, unknown fields, and formatting.
• Codex codex_home path: session listing, history, and deletion now consistently use the configured codex_home instead of hardcoded ~/.codex.
• Feishu card callback hint: log a reminder when interactive card mode is enabled but card.action.trigger may not be subscribed.

Added

• Weibo image & file support: send and receive images and files in Weibo DMs via base64 encoding within the WebSocket send_message payload. Implements ImageSender and FileSender interfaces.
• Comprehensive session tests: 12 new SessionManager unit tests covering PastAgentSessionIDs, legacy data migration, and version-based schema detection. 9 new Engine integration tests covering /list visibility across /new, provider switch, and real-world legacy data scenarios, plus end-to-end session name mapping tests for all three agent ID patterns (immediate, EventText, EventResult).
• Config preservation tests: 8 new tests verifying comment and field preservation for SaveActiveProvider, SaveAgentModel, SaveProviderModel, SaveLanguage, SaveDisplayConfig, SaveTTSMode, multi-project config, and global provider refs.

First stable release of the 1.3 series. 555 commits since v1.2.1 with major new features, platform improvements, and broad community contributions.

Highlights

• Web Admin UI — Full management dashboard embedded in the binary via go:embed. Project CRUD, session monitoring, cron editor, provider management, chat interface, and i18n (en/zh/zh-TW/ja/es). Use cc-connect web to open directly in the browser with auto-login.
• Lifecycle Event Hooks — New [[hooks]] config to trigger shell commands or HTTP webhooks on 7 event types: message.received, message.sent, session.started, session.ended, cron.triggered, permission.requested, error. Async by default, fail-open, non-blocking.
• Skill Management — New /skills page in the web UI with local skill browser (per-project, per-agent) and recommended skill presets fetched from remote.
• Global Provider Management — Add, edit, delete providers in the web UI; import from cc-switch config; per-agent-type provider presets with featured/star badges.

New Features

• cc-connect web CLI command: auto-configure web admin, open browser with token-based login
• Feishu: auto-resolve @name mentions to clickable at-tags (resolve_mentions config)
• Feishu: multi-level reply chain recognition; done-emoji reaction after streaming
• Feishu: configurable progress display styles (compact/card)
• Claude Code: support CLI wrappers via cli_path; /effort command for reasoning effort; auto permission mode; disallowed_tools config
• Codex: runtime reply footer; preserve workspace app-server options
• Kimi CLI: new agent support
• Pi: new agent support
• Discord: preserve table formatting; proxy support; @everyone/@here broadcast
• Telegram: forum topic support; markdown table monospace rendering; command menu adaptation
• WeCom: configurable api_base_url for private deployments; file receiving via HTTP callback
• Weixin (ilink): personal chat platform with CDN media, QR setup, image/file/audio send
• Config: support ${ENV_VAR} placeholders in TOML values
• Core: /workspace init with local directory paths; /dir directory history; agent-sid command; auto-compress context on token threshold; outgoing rate limiting
• Daemon: preserve proxy env in systemd service

Bug Fixes

• Fix Windows cross-compilation (duplicate runas stub file)
• Fix web footer double 'v' prefix in version display
• Fix web modal overlay not covering full viewport (portal rendering)
• Fix provider preset cards: action buttons pinned to card bottom
• Fix web page content overlapping footer (global layout restructure)
• Fix Gemini image handling: save to workspace, prompt-based file references
• Fix Claude Code: unblock readLoop when child subprocesses hold stdout pipe
• Fix Codex: multiline prompt on resume; force-kill process group on stop
• Fix core: race condition during session cleanup; follow symlinked skill directories; persist agent_session_id; filter /list to cc-connect owned sessions
• Fix Feishu: slash commands in thread/reply context; user/chat name resolution in async goroutine
• Fix Telegram: UTF-8-safe command menu descriptions
• Fix TTS: don't send empty language_type to Qwen TTS API
• Fix config: formatTOML no longer strips user-set zero values
• Security: mask bridge token in /api/v1/status; path traversal protection for static files

Contributors

Thanks to all contributors who made this release possible:

• @leoliang1997 — Feishu card rendering, auto-resolve @mentions
• @xukp20 — Provider env handling, skill discovery, Codex options
• @boyu-zhu — Telegram markdown table rendering
• @RukawaKaede — Claude Code CLI wrapper support
• @meishaoqing — Feishu multi-level reply chain
• @Zx55 — Telegram command menu, symlinked skill dirs
• @leighstillard — Claude Code /effort command
• @ht290 — inject_sender display name
• @Sentixxx — Claude Code readLoop subprocess fix
• @bugwz — WeCom private deployment API base URL
• @cold2600438-lgtm — Kimi CLI agent
• @MeteorSkyOne — Discord table formatting
• @happyTonakai — Feishu done-emoji reaction
• @xxb — Codex reply footer, Discord session routing
• @q107580018 — Feishu delete/model card flows
• @Cigarrr — Workspace binding parsing
• @g1f9 — Local directory workspace init
• @0xsegfaulted — agent-sid command
• @yzlu0917 — Env var config placeholders
• @sidney061212-ai — Agent session ID persistence
• @zkunzhu — Daemon proxy env preservation
• @Yuri0314 — TTS language type fix