qq-ai-bot

• Public OneBot ecosystem placement — qq-ai-bot now appears in the OneBot ecosystem application cases list instead of living only on its own README
• Why that matters — this gives the repo one third-party ecosystem anchor that is native to the QQ / OneBot audience, not just generic AI-awesome lists
• Official OneBot discussion is live — the repo is now also posted in the official OneBot “应用 / SDK” discussions, so people browsing the protocol community itself can discover it without leaving GitHub
• NapCat community thread is live — the repo is now also posted in NapCatQQ Show and tell so operators already in that ecosystem can discover it without leaving GitHub
• NapCat docs integration landed — qq-ai-bot is now also documented in NapCat community resources / integration docs, which is stronger than only having a showcase thread
• LLOneBot docs PR is open — qq-ai-bot is now also proposed for the official LLOneBot docs as a framework integration path
• Docker Compose sample PR is open — a concrete NapCat + qq-ai-bot sample PR is now proposed to Docker's Compose samples repo, which is a better fit than generic awesome-list spray
• AwesomeHomelab PR is open — qq-ai-bot is now also proposed to AwesomeHomelab, which is a cleaner homelab-facing discovery path than another generic AI list
• CasaOS app-store PR is open — qq-ai-bot is now also proposed to CasaOS-AppStore-Play as a NapCat WebUI + bundled bridge stack, which is the cleanest current path toward real homelab installs
• Chinese discovery surfaces are live — the repo is now also circulating through public recommendation / weekly threads such as HelloGitHub, GitHubDaily, 阮一峰周刊, 老胡周刊, and developer-plus weekly, so discovery is no longer only GitHub-native protocol communities
• Stable + moving image paths are live — the repo now ships both ghcr.io/happysnaker/qq-ai-bot:v0.1.7 and ghcr.io/happysnaker/qq-ai-bot:latest, so self-hosted directories can point to a stable version while operators who want the newest default-branch build still have a moving tag
• Multi-arch image is live — both public image paths now expose linux/amd64 and linux/arm64, with arm64 QEMU smoke automation keeping the homelab / NAS / SBC install story honest while real physical-host reports collect in issue #26
• ACP protocol discussion is live — a protocol-facing question based on qq-ai-bot's richer-media boundary is now posted in the official ACP discussions
• ACP docs listing merged — qq-ai-bot is now included in the official ACP clients documentation under Messaging
• Operational honesty — the public deployment matrix now separates “repo-owned stronger paths” from “documented but not overclaimed” paths, which makes the project feel more serious

• Community proof threads — Show and tell and deployment matrix Q&A are open so real deployment reports can accumulate in public
• LLOneBot community thread is live — there is now a matching LLOneBot Show and tell post for the users who operate closer to that stack
• Best short pitch — OneBot 11 + ACP + persistent sessions + progress streaming + Docker demo

• Already proven: `v0.1.7` image, Docker publish, CI, CodeQL, and arm64 QEMU smoke are green.
• Still needed: physical ARM host or CasaOS app-store flow with redacted evidence.
• Good testers: Raspberry Pi, Orange Pi, ARM NAS, CasaOS host, or real `linux/arm64` Docker host.
• Support note: use qq-ai-bot #26 arm64 if you want to fund this validation instead of testing it yourself.

• 15-minute tester pack
• Tracking issue #26
• Call for testers discussion
• Arm64 QEMU smoke workflow

• keep QQ transport logic independent from any specific model vendor
• reuse an ACP-style local agent bridge instead of hard-wiring one chatbot backend
• preserve session continuity per private chat or group conversation
• stream plan / tool / progress events back into QQ without inventing a custom protocol

• make the repo feel like a real bot infrastructure scaffold, not a fragile demo
• leave room for Telegram / WeChat / Feishu-style channel adapters later
• support local self-hosted agent workflows and operationally cleaner deployments
• give OneBot + ACP users a faster starting point than wiring everything from scratch

• v0.1.0 released — public release with OneBot 11 transport, ACP bridge, Docker demo, and observability surfaces
• v0.1.1 released — patch release that adds lightweight inbound replay dedupe and exposes duplicate-drop metrics
• v0.1.2 released — adds per-interaction correlation IDs across receive / dedupe / ACP dispatch / progress / reply so one noisy turn is easier to trace end to end
• v0.1.3 released — stages richer-media fallback handling so file / audio / video-like attachments are no longer silently ignored before the agent sees any warning
• v0.1.4 released — stabilizes the public Docker image tag for directories and homelab writeups
• v0.1.5 released — adds latency histograms for stronger runtime observability
• v0.1.6 released — adds a Postgres-backed session-store option for longer-running deployments
• v0.1.7 released — adds the public arm64 image-smoke workflow and verifies the tag image through GitHub Actions
• Deployment matrix published — public docs now separate stronger reverse-path defaults from forward-mode / platform combinations that are only documented support
• New FAQ + Discussions flow — common setup mistakes now route through FAQ and Q&A instead of vague issues
• DeepSeek-ready positioning — public docs now explicitly describe the common DeepSeek-backed local-agent deployment path
• Show-and-tell thread is live — community deployments can now collect in one public place instead of being scattered in ad-hoc comments
• Replay guard landed — the bridge now has a lightweight inbound dedupe layer to reduce obvious OneBot replays before they become duplicate ACP calls

• Active ecosystem outreach — now includes merged NapCat and ACP docs integrations, plus still-open Docker Compose and LLOneBot ecosystem PRs
• OneBot ecosystem listed — already included on the public OneBot ecosystem application cases page
• ACP ecosystem listed — already merged into the public awesome-agent-client-protocol list
• NapCat docs integration merged — qq-ai-bot now appears in NapCat community docs and integration guidance
• Docker Compose sample PR opened — a NapCat + qq-ai-bot stack is now proposed to docker/awesome-compose
• LLOneBot docs PR updated — the qq-ai-bot integration guide PR for official LLOneBot docs now also clarifies the LLOneBot forward-WS root-path detail
• ACP protocol discussion posted — a protocol-facing richer-media question is now live in the ACP community
• ACP official docs listing merged — qq-ai-bot is now included on the official ACP clients page
• OneBot official discussion posted — live in the official OneBot “应用 / SDK” discussions
• NapCat show-and-tell thread posted — live in NapCatQQ discussions
• LLOneBot show-and-tell thread posted — live in LLOneBot discussions
• AwesomeHomelab PR opened — a homelab-facing listing PR is now open so discovery is not limited to protocol-native communities
• Three claimable issues are public — latency histograms, Postgres session store, and multi-instance deployment notes are now open instead of leaving the repo looking announcement-only
• New self-hosted AI listing live — now proposed to an actively maintained AI agents ecosystem list under Self-Hosted Agents and UIs
• Best short pitch — OneBot 11 + ACP + persistent sessions + progress streaming + Docker demo
• Latest engineering signal — unsupported richer media is now surfaced explicitly to the agent instead of being silently dropped, which is a small but real runtime hardening step
• Latest operations signal — deployment paths now have a public confidence matrix instead of one vague “supports NapCat / LLOneBot” claim
• Support path stays simple — if this repo saved you time, ¥9.9 / ¥19.9 is still the lowest-friction thank-you path

• released as v0.1.0 with CI, Docker publishing, and build metadata now exposed in status endpoints
• v0.1.1 adds a replay-dedupe guard and makes duplicate-drop behavior visible in metrics
• v0.1.2 adds per-interaction correlation IDs so operators can grep one user turn across receive, dedupe, progress, and reply logs
• v0.1.3 makes richer-media limitations honest in runtime behavior by surfacing unsupported file / audio / video segments to the prompt layer
• v0.1.5 adds latency histograms, v0.1.6 adds Postgres session storage, and v0.1.7 adds public arm64 image-smoke evidence
• deployment-matrix.md now says explicitly which routes are stronger repo-owned paths and which are only documented compatibility paths
• richer-media stage 1 now makes the repo safer to demo because unsupported attachments no longer create the illusion that the agent already consumed them
• /metrics now exposes Prometheus-style runtime counters for connection state, traffic, ACP calls, and conversation gauges
• has a public roadmap instead of pretending the current shape is the final shape
• actively invites contributions on session storage, richer media, and stronger runtime visibility

• better than a throwaway demo: it already has architecture notes, quickstart docs, and health endpoints
• best next high-signal OSS work here is Redis session store, richer media, and tracing / deeper observability
• if you want your own bot repo packaged this clearly, the blunt paid read starts at ¥29.9

• OneBot 11 reverse and forward WebSocket modes
• private chat + group mention triggers
• persistent conversation sessions keyed by QQ conversation
• image download and pass-through into the ACP agent flow
• progress message streaming from plan / tool events

• health and readiness endpoints
• mock ACP agent for local integration work
• ESLint, typecheck, tests, and GitHub Actions
• Docker packaging for cleaner deployment
• clear extension path for richer media and stronger observability

• engineers building self-hosted QQ bot infrastructure
• developers experimenting with ACP-compatible local coding / chat agents
• people who want a better starting point than mixing transport code and model code together

• maintainers who want a clearer architecture before adding Redis, tracing, or moderation logic
• hackers exploring multi-channel bot infrastructure after starting from QQ / OneBot
• anyone who wants a modern repo that feels more like middleware / orchestration than a toy bot

• ¥9.9 / ¥19.9 — thank-you tip if this repo saved you setup or wiring time
• Best payment note — qq-ai-bot or qq-ai-bot setup
• Fastest path — scan WeChat Pay / Alipay on the support page
• ¥29.9 deploy read — best if your bot technically runs but the token / WS / Docker / landing-page story is still muddy

• ¥29.9 quick read — one blunt paragraph + top 3 fixes for your own bot / README / landing page
• ¥99 async review — GitHub profile, pinned repos, README cleanup, or one focused landing page
• Public intake guardrail — do not paste private logs, credentials, QR codes, payment screenshots, internal URLs, or raw live integration output in public issues; use sponsor / paid-support intake replies first
• Route by audience — use the sponsor prospect pipeline to choose the right proof, CTA, support note, and guardrail for homelab testers, bot builders, curators, sponsors, and paid-review customers; then use the sponsor conversion scorecard to decide Hot / Warm / Nurture / No-send before posting
• Best fit — if you want your own AI / infra repo to feel less toy-like and more hireable

Shortest line

OneBot 11 + ACP + persistent sessions + progress streaming + Docker demo

Chinese version

qq-ai-bot 是一个面向 OneBot 11 / NapCat / LLOneBot 的 QQ ↔ AI 自托管脚手架。它把 QQ 消息入口、ACP agent bridge、会话持久化、进度回传和 Docker demo 整理成了一个更像 bot 基础设施的仓库，而不是只会聊天的玩具 demo。

English version

qq-ai-bot is a self-hosted QQ ↔ AI bot scaffold for OneBot 11 / NapCat / LLOneBot and ACP-compatible agents, with persistent sessions, progress streaming, and a Docker demo. It is positioned as bot infrastructure / integration glue rather than another toy chat UI.

Homelab / CasaOS tester ask

Need one real ARM/CasaOS tester for qq-ai-bot.

CI/QEMU arm64 smoke is green, but I still need a physical ARM/NAS/SBC/CasaOS report.

Tester pack:
https://github.com/happysnaker/qq-ai-bot/blob/main/docs/public/arm64-casaos-tester-pack.md

Outreach kit:
https://github.com/happysnaker/qq-ai-bot/blob/main/docs/public/homelab-outreach-kit.md

Best 3-link pack

Repo: https://github.com/happysnaker/qq-ai-bot
Project page: https://happysnaker.github.io/qq-ai-bot/
Outreach kit: https://github.com/happysnaker/qq-ai-bot/blob/main/docs/public/homelab-outreach-kit.md