Add ProcessingStatusNotifier implementation: show 👀 reaction while
processing, remove on completion/failure. Fix isReplyToBot to match
only the current bot (bot.Self.ID) instead of any bot, preventing
multiple bots from responding to the same reply. Add rune-safe text
truncation for Telegram message length limit.
Strip invalid UTF-8 byte sequences in sendTelegramTextReturnMessage and
editTelegramMessageText to prevent "text must be encoded in UTF-8" errors
that abort the stream mid-response.
Catch unique constraint violation (SQLSTATE 23505) during route creation
in ResolveConversation and fall back to Find, preventing duplicate key
errors when concurrent inbound messages hit the same chat simultaneously.
Add shared IsUniqueViolation helper to internal/db.
Increase edit throttle from 250ms to 1500ms to respect Telegram per-chat limits,
remove newline bypass that effectively disabled throttling, add dedicated
editStreamMessageFinal with retry for final message delivery, and simplify
editTelegramMessageText by removing blocking sleep in favor of caller-level retry.
- Treat 400 "message is not modified" as success to avoid user-facing error
- On 429: sleep retry_after and retry once in editTelegramMessageText;
stream backs off lastEditedAt and returns nil
- Require error code 400 for message-not-modified check; add production
error string to unit tests
- Lower base throttle to 250ms; add test hooks and tests for 429 retry
and stream backoff
- README: emphasize Docker requirement with get-started link, move silent
install under one-click and de-emphasize
- DEPLOYMENT: clarify default config path and that only config file is mounted
- Move CONTRIBUTING.md to .github/CONTRIBUTING.md (git mv)
Tool call messages disappeared after page refresh because messageToChat()
filtered out messages without text content. Added convertMessagesToChats()
to merge consecutive assistant(tool_calls) + tool(results) + assistant(text)
into a single ChatMessage with ToolCallBlocks.
Propagate conversation type (direct/group/thread) from channel adapters
all the way to the agent prompt. Store conversation_type on bot_channel_routes
so the bot knows whether a message originates from a p2p chat, group, or thread.
Schema changes are folded into the 0001 init migration (destructive update).
- Refactor RuntimeChecker interface: CheckKeys() + RunCheck() for
individual check dispatch instead of batch-all
- Add GET /bots/:id/checks/keys to list all available check keys
- Add GET /bots/:id/checks/run/:key to evaluate a single check
- MCP ConnectionChecker probes each active connection independently
via tools/list with 8s timeout
- Keep container checks (init/record/task/data_path) as fast builtins
- Graceful network setup failure in containerd handler (log warning
instead of killing task) for containerd-in-docker compatibility
- Add RuntimeChecker interface for extensible bot health checks
- Implement MCP ConnectionChecker: probes active connections via tools/list
- Show MCP server status (healthy/unreachable/no tools) in bot checks
- Register checker in main.go alongside container lifecycle
- Add pid: host to containerd service so server can access MCP container
network namespaces via /proc/PID/ns/net for CNI setup
- Add Node.js, npm, Python 3, uv to embedded MCP image rootfs so users
can run npx/uvx MCP servers inside containers
In containerd-in-docker mode, SetupNetwork fails because netns is
unavailable. Previously this killed the task, making stdio MCP tools
unusable. Now the task continues running with a warning log, since
stdio MCP communication does not require networking.
- Add bot-mcp.vue component with table, create/edit dialog, import/export
- Wire MCP tab into bot detail page
- Remove dead global MCP page, composable, and form component (/mcp route)
- Add i18n keys for import/export/copy and MCP CRUD messages
- Fix store/User.ts case sensitivity for Linux builds
- Accept standard mcpServers item format (command/args/env/url/headers)
- Auto-infer connection type: command -> stdio, url -> http/sse
- Add PUT /bots/:bot_id/mcp/import for batch import from mcpServers dict
- Add GET /bots/:bot_id/mcp/export for standard format export
- Add UpsertMCPConnectionByName SQL for import upsert by name
- Preserve is_active state on import upsert
Normalize JSON tags to snake_case, remove dead fusion branches and unused
struct fields, add proper error logging for BM25 operations, wire AgentID
through filters/store/retrieval, and replace goto with structured control flow.
Run containerd as a sidecar container instead of requiring host
containerd installation. Server mounts shared docker volumes for
containerd socket, data and state.
Acbox