🐈 nanobot v0.1.4.post3 just dropped — 33 PRs merged, 16 new contributors! You guys are unstoppable 🔥

This release is all about making the agent smarter by seeing less junk: cleaner context, hardened session history, and fewer ghost messages. Less noise in → less hallucination out. Plus a shiny new Matrix channel and experimental thinking mode.

Highlights

• Agent Loop Hardening — Empty assistant messages (no content, no tool calls) are now filtered instead of poisoning context; LLM error responses are no longer saved to history, preventing permanent 400-error loops; message tool suppression scoped to same-channel only (#1314, #1198, #1206)
• Context Noise Reduction — Runtime metadata (time, channel, chat ID) separated into an untrusted layer and excluded from session history, so the agent sees only what matters — fewer tokens, less confusion, less hallucination (#1126, #1222)
• Session Safety — Base64 images stripped from history to prevent context overflow; null responses no longer corrupt future turns (#1191, #1314)
• Matrix Channel — Full Matrix (Element) chat channel support with E2EE, media uploads, and typing indicators (#420, #1239)
• Provider Compatibility — Short alphanumeric tool_call_id for Mistral; list-type tool arguments handled gracefully; explicit provider selection in config (#1293, #1294, #1214, #1316)
• Thinking Mode (experimental) — New reasoning_effort config enables LLM reasoning for supported models; session history preserves reasoning_content and thinking_blocks across turns (#1351, #1330, #1074)
• Subagent Improvements — /stop cancels spawned subagents; streamlined prompt eliminates dead code and reuses ContextBuilder (#1180, #1347)
• Feishu — Interactive card text extraction fix, configurable reaction emoji, corrected bot permissions (#1323, #1257, #1317, #1348)
• DingTalk — Images and media sent as proper message types instead of plain-text links (#1337)
• Telegram — Media groups aggregated into a single inbound message (#1258)
• QQ — Fixed C2C reply permissions, disabled file log on read-only filesystems (#1307, #1346)
• WhatsApp — Message deduplication prevents redundant agent processing (#1325)
• Shell & Security — Full Windows path parsing in workspace guard; configurable path_append for subprocess PATH (#1286, #1083)
• Memory & Concurrency — WeakValueDictionary for consolidation locks eliminates race conditions and manual cleanup (#1326)
• Workspace — Automatic template synchronization restores critical files on startup (#1253)

What's Changed

• fix: preserve reasoning_content in messages for thinking-enabled models by @haosenwang1018 in https://github.com/HKUDS/nanobot/pull/1074
• feat/fix(exec): add path_append config to extend PATH for subprocess by @aiguozhi123456 in https://github.com/HKUDS/nanobot/pull/1083
• feat: add stable system layer + untrusted runtime context layer by @pikaxinge in https://github.com/HKUDS/nanobot/pull/1126
• feat: /stop cancels spawned subagents via session tracking by @coldxiangyu163 in https://github.com/HKUDS/nanobot/pull/1180
• feat: support explicit provider selection in config by @Re-bin in https://github.com/HKUDS/nanobot/pull/1214
• Fix assistant messages without tool calls not being saved to session history by @VITOHJL in https://github.com/HKUDS/nanobot/pull/1198
• Fix: The base64 images are stored in the session history, causing context overflow. by @dxtime in https://github.com/HKUDS/nanobot/pull/1191
• feat: add Matrix (Element) chat channel support by @tanishra in https://github.com/HKUDS/nanobot/pull/420
• fix(agent): only suppress final reply when message tool sends to same… by @chengyongru in https://github.com/HKUDS/nanobot/pull/1206
• fix(web): use self.api_key instead of undefined api_key by @yongPhone in https://github.com/HKUDS/nanobot/pull/1228
• fix(telegram): aggregate media groups into a single inbound message by @KimGLee in https://github.com/HKUDS/nanobot/pull/1258
• feat(feishu): make reaction emoji configurable by @kimkitsuragi26 in https://github.com/HKUDS/nanobot/pull/1257
• feat: automatic workspace template synchronization by @honjiaxuan in https://github.com/HKUDS/nanobot/pull/1253
• Fix Matrix channel initialization and configuration by @tanishra in https://github.com/HKUDS/nanobot/pull/1239
• fix(agent): avoid persisting runtime context metadata into history by @KimGLee in https://github.com/HKUDS/nanobot/pull/1222
• fix: update heartbeat tests to match two-phase tool-call architecture by @intelliot in https://github.com/HKUDS/nanobot/pull/1200
• Fix(prompt): guide llm grep using timestamp by @aiguozhi123456 in https://github.com/HKUDS/nanobot/pull/1278
• fix: generate short alphanumeric tool_call_id for Mistral compatibility by @Re-bin in https://github.com/HKUDS/nanobot/pull/1293
• fix: remove overly broad "codex" keyword from openai_codex provider by @nikolasdehor in https://github.com/HKUDS/nanobot/pull/1316
• fix: prevent session poisoning from null/error LLM responses by @nikolasdehor in https://github.com/HKUDS/nanobot/pull/1314
• fix the feishu interactive message text cannot be extracted by @azhengzz in https://github.com/HKUDS/nanobot/pull/1323
• Modify Feishu bot permissions in README by @yankeguo in https://github.com/HKUDS/nanobot/pull/1317
• fix: pass msg_id in QQ C2C reply to avoid proactive message permissio… by @GabrielWithTina in https://github.com/HKUDS/nanobot/pull/1307
• fix: handle list-type tool arguments in _tool_hint by @Michael-lhh in https://github.com/HKUDS/nanobot/pull/1294
• fix(shell): parse full Windows absolute paths in workspace guard by @fengxiaohu in https://github.com/HKUDS/nanobot/pull/1286
• refactor: use WeakValueDictionary for consolidation locks by @Re-bin in https://github.com/HKUDS/nanobot/pull/1326
• fix: add message deduplication to WhatsApp channel by @spartan077 in https://github.com/HKUDS/nanobot/pull/1325
• feat(dingtalk): send images as image messages, keep files as attachments by @asdf17128 in https://github.com/HKUDS/nanobot/pull/1337
• refactor: streamline subagent prompt by reusing ContextBuilder and SkillsLoader by @Re-bin in https://github.com/HKUDS/nanobot/pull/1347
• fix(qq): disable botpy file log to fix read-only filesystem error by @zerone0x in https://github.com/HKUDS/nanobot/pull/1346
• feat: add reasoning_effort config to enable LLM thinking mode by @Re-bin in https://github.com/HKUDS/nanobot/pull/1351
• fix(lark): Remove non-existent stop() call on Lark ws.Client when enable lark channel by @yorkhellen in https://github.com/HKUDS/nanobot/pull/1348
• fix(providers): allow reasoning_content in message history by @fengxiaohu in https://github.com/HKUDS/nanobot/pull/1330

New Contributors

• @aiguozhi123456 made their first contribution in https://github.com/HKUDS/nanobot/pull/1083
• @VITOHJL made their first contribution in https://github.com/HKUDS/nanobot/pull/1198
• @tanishra made their first contribution in https://github.com/HKUDS/nanobot/pull/420
• @yongPhone made their first contribution in https://github.com/HKUDS/nanobot/pull/1228
• @kimkitsuragi26 made their first contribution in https://github.com/HKUDS/nanobot/pull/1257
• @honjiaxuan made their first contribution in https://github.com/HKUDS/nanobot/pull/1253
• @intelliot made their first contribution in https://github.com/HKUDS/nanobot/pull/1200
• @azhengzz made their first contribution in https://github.com/HKUDS/nanobot/pull/1323
• @yankeguo made their first contribution in https://github.com/HKUDS/nanobot/pull/1317
• @GabrielWithTina made their first contribution in https://github.com/HKUDS/nanobot/pull/1307
• @Michael-lhh made their first contribution in https://github.com/HKUDS/nanobot/pull/1294
• @fengxiaohu made their first contribution in https://github.com/HKUDS/nanobot/pull/1286
• @spartan077 made their first contribution in https://github.com/HKUDS/nanobot/pull/1325
• @asdf17128 made their first contribution in https://github.com/HKUDS/nanobot/pull/1337
• @zerone0x made their first contribution in https://github.com/HKUDS/nanobot/pull/1346
• @yorkhellen made their first contribution in https://github.com/HKUDS/nanobot/pull/1348

Full Changelog: https://github.com/HKUDS/nanobot/compare/v0.1.4.post2...v0.1.4.post3