Merge remote-tracking branch 'origin/main' into feat/echo-voice-audio-transcription

# Conflicts:
#	pkg/channels/telegram/telegram.go
This commit is contained in:
afjcjsbx
2026-03-07 18:55:32 +01:00
18 changed files with 1132 additions and 53 deletions
+1
View File
@@ -16,6 +16,7 @@ import (
_ "github.com/sipeed/picoclaw/pkg/channels/dingtalk"
_ "github.com/sipeed/picoclaw/pkg/channels/discord"
_ "github.com/sipeed/picoclaw/pkg/channels/feishu"
_ "github.com/sipeed/picoclaw/pkg/channels/irc"
_ "github.com/sipeed/picoclaw/pkg/channels/line"
_ "github.com/sipeed/picoclaw/pkg/channels/maixcam"
_ "github.com/sipeed/picoclaw/pkg/channels/onebot"
+22
View File
@@ -164,6 +164,28 @@
"max_steps": 10,
"welcome_message": "Hello! I'm your AI assistant. How can I help you today?",
"reasoning_channel_id": ""
},
"irc": {
"enabled": false,
"server": "irc.libera.chat:6697",
"tls": true,
"nick": "mybot",
"user": "",
"real_name": "",
"password": "",
"nickserv_password": "",
"sasl_user": "",
"sasl_password": "",
"channels": ["#mychannel"],
"request_caps": ["server-time", "message-tags"],
"allow_from": [],
"group_trigger": {
"mention_only": true
},
"typing": {
"enabled": false
},
"reasoning_channel_id": ""
}
},
"providers": {
+145
View File
@@ -0,0 +1,145 @@
# Agent Refactor
## What this directory is for
This directory is the working area for the current Agent refactor.
The purpose of this refactor is simple:
the project needs a smaller, clearer, and more stable Agent model before more Agent-related behavior is added.
The codebase already contains meaningful Agent behavior. What it still lacks is a sufficiently explicit and stable semantic boundary around that behavior.
This refactor exists to fix that first.
---
## Refactor stance
This is a maintenance-led consolidation effort.
It is not a general invitation to expand Agent behavior in parallel.
During this refactor window, Agent-related work should converge on the current refactor track instead of branching into new semantics.
That means:
- concept clarification before feature expansion
- boundary tightening before abstraction growth
- semantic consolidation before new behavior
---
## Core rule: minimum concepts only
This refactor follows one hard rule:
**do not introduce a new concept unless it is strictly necessary**
More explicitly:
- if an existing concept can be clarified, reuse it
- if an existing boundary can be made explicit, do that first
- if a behavior can be expressed without a new abstraction, do not add one
- "future flexibility" is not enough justification on its own
The goal of this refactor is not to grow the model.
The goal is to reduce ambiguity.
---
## What is being clarified
This refactor is currently concerned with the following questions:
1. what an `Agent` is
2. what an `AgentLoop` is
3. what the lifecycle of `AgentLoop` is
4. what the event surface around `AgentLoop` is
5. how persona / identity is assembled
6. how capabilities are represented
7. how context boundaries and compression work
8. how subagent coordination works
These are the current working boundaries.
If they need to be adjusted, they should be adjusted explicitly rather than drift implicitly in code.
---
## Status of this directory
The documents here are working materials.
They are not final or immutable.
If current notes are incomplete, incorrectly split, or too broad, they should be revised. This directory should evolve with the refactor rather than pretending the first draft is complete.
---
## Suggested document split
This directory may eventually contain notes such as:
- `agent-overview.md`
- what an Agent is
- `agent-loop.md`
- AgentLoop contract, lifecycle, event surface
- `persona.md`
- persona and identity assembly
- `capability.md`
- tools / skills / MCP capability semantics
- `context.md`
- context scope, history, summary, compression
- `subagent.md`
- subagent coordination rules
These files should be added only when they help clarify the current refactor work.
This directory should not turn into a generic architecture dump.
---
## What this directory is not for
This directory is not intended for:
- broad speculative architecture
- future multi-node protocol design not required by the current refactor
- parallel feature planning unrelated to Agent consolidation
- adding new concepts before current ones are made clear
If a topic does not directly help reduce ambiguity in the current Agent model, it probably does not belong here yet.
---
## Relationship to implementation
Implementation changes should not keep redefining Agent semantics implicitly.
If a PR changes or depends on Agent semantics, those semantics should either already exist here or be clarified in a linked issue first.
This directory is here to make implementation narrower and more disciplined.
---
## Relationship to GitHub tracking
The umbrella issue for this refactor should point here.
The issue is the coordination surface.
This directory is the repository-local working surface.
---
## Summary
The main question of this refactor is not:
- what more can Agent do
The main question is:
- what is the smallest stable model that current Agent behavior can be organized around
+1
View File
@@ -37,6 +37,7 @@ require (
github.com/davecgh/go-spew v1.1.1 // indirect
github.com/dustin/go-humanize v1.0.1 // indirect
github.com/elliotchance/orderedmap/v3 v3.1.0 // indirect
github.com/ergochat/irc-go v0.5.0 // indirect
github.com/gdamore/encoding v1.0.1 // indirect
github.com/inconshreveable/mousetrap v1.1.0 // indirect
github.com/lucasb-eyer/go-colorful v1.3.0 // indirect
+2
View File
@@ -48,6 +48,8 @@ github.com/dustin/go-humanize v1.0.1 h1:GzkhY7T5VNhEkwH0PVJgjz+fX1rhBrR7pRT3mDkp
github.com/dustin/go-humanize v1.0.1/go.mod h1:Mu1zIs6XwVuF/gI1OepvI0qD18qycQx+mFykh5fBlto=
github.com/elliotchance/orderedmap/v3 v3.1.0 h1:j4DJ5ObEmMBt/lcwIecKcoRxIQUEnw0L804lXYDt/pg=
github.com/elliotchance/orderedmap/v3 v3.1.0/go.mod h1:G+Hc2RwaZvJMcS4JpGCOyViCnGeKf0bTYCGTO4uhjSo=
github.com/ergochat/irc-go v0.5.0 h1:woQ1RS9YbfgqPgSpPBBQeczXGIGzR0aC7dEgk469fTw=
github.com/ergochat/irc-go v0.5.0/go.mod h1:2vi7KNpIPWnReB5hmLpl92eMywQvuIeIIGdt/FQCph0=
github.com/fsnotify/fsnotify v1.4.7/go.mod h1:jwhsz4b93w/PPRr/qN1Yymfu8t87LnFCMoQvtojpjFo=
github.com/fsnotify/fsnotify v1.4.9/go.mod h1:znqG4EE+3YCdAaPaxE2ZRY/06pZUdp0tY4IgpuI1SZQ=
github.com/gdamore/encoding v1.0.1 h1:YzKZckdBL6jVt2Gc+5p82qhrGiqMdG/eNs6Wy0u3Uhw=
+45 -11
View File
@@ -220,7 +220,7 @@ func registerSharedTools(
// Spawn tool with allowlist checker
if cfg.Tools.IsToolEnabled("spawn") {
if cfg.Tools.IsToolEnabled("subagent") {
subagentManager := tools.NewSubagentManager(provider, agent.Model, agent.Workspace, msgBus)
subagentManager := tools.NewSubagentManager(provider, agent.Model, agent.Workspace)
subagentManager.SetLLMOptions(agent.MaxTokens, agent.Temperature)
spawnTool := tools.NewSpawnTool(subagentManager)
currentAgentID := agentID
@@ -759,9 +759,8 @@ func (al *AgentLoop) runAgentLoop(
agent *AgentInstance,
opts processOptions,
) (string, error) {
// 0. Record last channel for heartbeat notifications (skip internal channels)
// 0. Record last channel for heartbeat notifications (skip internal channels and cli)
if opts.Channel != "" && opts.ChatID != "" {
// Don't record internal channels (cli, system, subagent)
if !constants.IsInternalChannel(opts.Channel) {
channelKey := fmt.Sprintf("%s:%s", opts.Channel, opts.ChatID)
if err := al.RecordLastChannel(channelKey); err != nil {
@@ -1098,9 +1097,12 @@ func (al *AgentLoop) runLLMIteration(
"target_channel": al.targetReasoningChannelID(opts.Channel),
"channel": opts.Channel,
})
// Check if no tool calls - we're done
// Check if no tool calls - then check reasoning content if any
if len(response.ToolCalls) == 0 {
finalContent = response.Content
if finalContent == "" && response.ReasoningContent != "" {
finalContent = response.ReasoningContent
}
logger.InfoCF("agent", "LLM response without tool calls (direct answer)",
map[string]any{
"agent_id": agent.ID,
@@ -1186,15 +1188,47 @@ func (al *AgentLoop) runLLMIteration(
"iteration": iteration,
})
// Create async callback for tools that implement AsyncExecutor
asyncCallback := func(callbackCtx context.Context, result *tools.ToolResult) {
// Create async callback for tools that implement AsyncExecutor.
// When the background work completes, this publishes the result
// as an inbound system message so processSystemMessage routes it
// back to the user via the normal agent loop.
asyncCallback := func(_ context.Context, result *tools.ToolResult) {
// Send ForUser content directly to the user (immediate feedback),
// mirroring the synchronous tool execution path.
if !result.Silent && result.ForUser != "" {
logger.InfoCF("agent", "Async tool completed, agent will handle notification",
map[string]any{
"tool": tc.Name,
"content_len": len(result.ForUser),
})
outCtx, outCancel := context.WithTimeout(context.Background(), 5*time.Second)
defer outCancel()
_ = al.bus.PublishOutbound(outCtx, bus.OutboundMessage{
Channel: opts.Channel,
ChatID: opts.ChatID,
Content: result.ForUser,
})
}
// Determine content for the agent loop (ForLLM or error).
content := result.ForLLM
if content == "" && result.Err != nil {
content = result.Err.Error()
}
if content == "" {
return
}
logger.InfoCF("agent", "Async tool completed, publishing result",
map[string]any{
"tool": tc.Name,
"content_len": len(content),
"channel": opts.Channel,
})
pubCtx, pubCancel := context.WithTimeout(context.Background(), 5*time.Second)
defer pubCancel()
_ = al.bus.PublishInbound(pubCtx, bus.InboundMessage{
Channel: "system",
SenderID: fmt.Sprintf("async:%s", tc.Name),
ChatID: fmt.Sprintf("%s:%s", opts.Channel, opts.ChatID),
Content: content,
})
}
toolResult := agent.Tools.ExecuteWithContext(
+154
View File
@@ -0,0 +1,154 @@
package irc
import (
"fmt"
"strings"
"time"
"unicode"
"github.com/ergochat/irc-go/ircevent"
"github.com/ergochat/irc-go/ircmsg"
"github.com/sipeed/picoclaw/pkg/bus"
"github.com/sipeed/picoclaw/pkg/identity"
"github.com/sipeed/picoclaw/pkg/logger"
)
// onConnect is called after a successful connection (and on reconnect).
func (c *IRCChannel) onConnect(conn *ircevent.Connection) {
// NickServ auth (only if SASL is not configured)
if c.config.NickServPassword != "" && c.config.SASLUser == "" {
conn.Privmsg("NickServ", "IDENTIFY "+c.config.NickServPassword)
}
// Join configured channels
for _, ch := range c.config.Channels {
conn.Join(ch)
logger.InfoCF("irc", "Joined IRC channel", map[string]any{
"channel": ch,
})
}
}
// onPrivmsg handles incoming PRIVMSG events.
func (c *IRCChannel) onPrivmsg(conn *ircevent.Connection, e ircmsg.Message) {
if len(e.Params) < 2 {
return
}
nick := e.Nick()
currentNick := conn.CurrentNick()
// Ignore own messages
if strings.EqualFold(nick, currentNick) {
return
}
target := e.Params[0] // channel name or bot's nick
content := e.Params[1] // message text
// Determine if this is a DM or channel message
isDM := !strings.HasPrefix(target, "#") && !strings.HasPrefix(target, "&")
var chatID string
var peer bus.Peer
if isDM {
chatID = nick
peer = bus.Peer{Kind: "direct", ID: nick}
} else {
chatID = target
peer = bus.Peer{Kind: "group", ID: target}
}
sender := bus.SenderInfo{
Platform: "irc",
PlatformID: nick,
CanonicalID: identity.BuildCanonicalID("irc", nick),
Username: nick,
DisplayName: nick,
}
if !c.IsAllowedSender(sender) {
return
}
// For channel messages, check group trigger (mention detection)
if !isDM {
isMentioned := isBotMentioned(content, currentNick)
if isMentioned {
content = stripBotMention(content, currentNick)
}
respond, cleaned := c.ShouldRespondInGroup(isMentioned, content)
if !respond {
return
}
content = cleaned
}
if strings.TrimSpace(content) == "" {
return
}
messageID := fmt.Sprintf("%s-%d", nick, time.Now().UnixNano())
metadata := map[string]string{
"platform": "irc",
"server": c.config.Server,
}
if !isDM {
metadata["channel"] = target
}
c.HandleMessage(c.ctx, peer, messageID, nick, chatID, content, nil, metadata, sender)
}
// nickMentionedAt returns the byte index where botNick is mentioned in content
// with word-boundary checks, or -1 if not found. Also checks for "nick:" /
// "nick," prefix convention.
func nickMentionedAt(content, botNick string) int {
lower := strings.ToLower(content)
lowerNick := strings.ToLower(botNick)
// "nick:" or "nick," at start (most common IRC convention)
if strings.HasPrefix(lower, lowerNick+":") || strings.HasPrefix(lower, lowerNick+",") {
return 0
}
// Word-boundary match anywhere in the message
idx := strings.Index(lower, lowerNick)
if idx < 0 {
return -1
}
runes := []rune(lower)
nickRunes := []rune(lowerNick)
endIdx := idx + len(string(nickRunes))
before := idx == 0 || !unicode.IsLetter(runes[idx-1]) && !unicode.IsDigit(runes[idx-1])
after := endIdx >= len(lower) || !unicode.IsLetter(rune(lower[endIdx])) && !unicode.IsDigit(rune(lower[endIdx]))
if before && after {
return idx
}
return -1
}
// isBotMentioned checks if the bot's nick appears in the message.
func isBotMentioned(content, botNick string) bool {
return nickMentionedAt(content, botNick) >= 0
}
// stripBotMention removes "nick: " or "nick, " prefix from content.
func stripBotMention(content, botNick string) string {
idx := nickMentionedAt(content, botNick)
if idx != 0 {
return content
}
lowerNick := strings.ToLower(botNick)
lower := strings.ToLower(content)
for _, sep := range []string{":", ","} {
prefix := lowerNick + sep
if strings.HasPrefix(lower, prefix) {
return strings.TrimSpace(content[len(prefix):])
}
}
return content
}
+16
View File
@@ -0,0 +1,16 @@
package irc
import (
"github.com/sipeed/picoclaw/pkg/bus"
"github.com/sipeed/picoclaw/pkg/channels"
"github.com/sipeed/picoclaw/pkg/config"
)
func init() {
channels.RegisterFactory("irc", func(cfg *config.Config, b *bus.MessageBus) (channels.Channel, error) {
if !cfg.Channels.IRC.Enabled {
return nil, nil
}
return NewIRCChannel(cfg.Channels.IRC, b)
})
}
+194
View File
@@ -0,0 +1,194 @@
package irc
import (
"context"
"crypto/tls"
"fmt"
"strings"
"github.com/ergochat/irc-go/ircevent"
"github.com/ergochat/irc-go/ircmsg"
"github.com/sipeed/picoclaw/pkg/bus"
"github.com/sipeed/picoclaw/pkg/channels"
"github.com/sipeed/picoclaw/pkg/config"
"github.com/sipeed/picoclaw/pkg/logger"
)
// IRCChannel implements the Channel interface for IRC servers.
type IRCChannel struct {
*channels.BaseChannel
config config.IRCConfig
conn *ircevent.Connection
ctx context.Context
cancel context.CancelFunc
}
// NewIRCChannel creates a new IRC channel.
func NewIRCChannel(cfg config.IRCConfig, messageBus *bus.MessageBus) (*IRCChannel, error) {
if cfg.Server == "" {
return nil, fmt.Errorf("irc server is required")
}
if cfg.Nick == "" {
return nil, fmt.Errorf("irc nick is required")
}
base := channels.NewBaseChannel("irc", cfg, messageBus, cfg.AllowFrom,
channels.WithMaxMessageLength(400),
channels.WithGroupTrigger(cfg.GroupTrigger),
channels.WithReasoningChannelID(cfg.ReasoningChannelID),
)
return &IRCChannel{
BaseChannel: base,
config: cfg,
}, nil
}
// Start connects to the IRC server and begins listening.
func (c *IRCChannel) Start(ctx context.Context) error {
logger.InfoC("irc", "Starting IRC channel")
c.ctx, c.cancel = context.WithCancel(ctx)
user := c.config.User
if user == "" {
user = c.config.Nick
}
realName := c.config.RealName
if realName == "" {
realName = c.config.Nick
}
caps := []string(c.config.RequestCaps)
if len(caps) == 0 {
caps = []string{"server-time", "message-tags"}
}
conn := &ircevent.Connection{
Server: c.config.Server,
Nick: c.config.Nick,
User: user,
RealName: realName,
Password: c.config.Password,
UseTLS: c.config.TLS,
RequestCaps: caps,
QuitMessage: "Goodbye",
Debug: false,
Log: nil,
}
if c.config.TLS {
conn.TLSConfig = &tls.Config{
ServerName: extractHost(c.config.Server),
}
}
// SASL auth (takes priority over NickServ)
if c.config.SASLUser != "" && c.config.SASLPassword != "" {
conn.SASLLogin = c.config.SASLUser
conn.SASLPassword = c.config.SASLPassword
}
// Register event handlers
conn.AddConnectCallback(func(e ircmsg.Message) {
c.onConnect(conn)
})
conn.AddCallback("PRIVMSG", func(e ircmsg.Message) {
c.onPrivmsg(conn, e)
})
if err := conn.Connect(); err != nil {
return fmt.Errorf("irc connect failed: %w", err)
}
c.conn = conn
// ircevent.Connection.Loop() handles reconnection internally.
go conn.Loop()
c.SetRunning(true)
logger.InfoCF("irc", "IRC channel started", map[string]any{
"server": c.config.Server,
"nick": c.config.Nick,
})
return nil
}
// Stop disconnects from the IRC server.
func (c *IRCChannel) Stop(ctx context.Context) error {
logger.InfoC("irc", "Stopping IRC channel")
c.SetRunning(false)
if c.conn != nil {
c.conn.Quit()
}
if c.cancel != nil {
c.cancel()
}
logger.InfoC("irc", "IRC channel stopped")
return nil
}
// Send sends a message to an IRC channel or user.
func (c *IRCChannel) Send(ctx context.Context, msg bus.OutboundMessage) error {
if !c.IsRunning() {
return channels.ErrNotRunning
}
target := msg.ChatID
if target == "" {
return fmt.Errorf("chat ID is empty: %w", channels.ErrSendFailed)
}
if strings.TrimSpace(msg.Content) == "" {
return nil
}
// Send each line separately (IRC is line-oriented)
lines := strings.Split(msg.Content, "\n")
for _, line := range lines {
line = strings.TrimRight(line, "\r")
if line == "" {
continue
}
c.conn.Privmsg(target, line)
}
logger.DebugCF("irc", "Message sent", map[string]any{
"target": target,
"lines": len(lines),
})
return nil
}
// StartTyping implements channels.TypingCapable using IRCv3 +typing client tag.
// Requires typing.enabled in config and server support for message-tags capability.
func (c *IRCChannel) StartTyping(ctx context.Context, chatID string) (func(), error) {
noop := func() {}
if !c.config.Typing.Enabled || !c.IsRunning() || c.conn == nil {
return noop, nil
}
// Check if server supports message-tags (required for TAGMSG)
if _, ok := c.conn.AcknowledgedCaps()["message-tags"]; !ok {
return noop, nil
}
c.conn.SendWithTags(map[string]string{"+typing": "active"}, "TAGMSG", chatID)
return func() {
if c.IsRunning() && c.conn != nil {
c.conn.SendWithTags(map[string]string{"+typing": "done"}, "TAGMSG", chatID)
}
}, nil
}
// extractHost returns the hostname portion of a host:port string.
func extractHost(server string) string {
host, _, found := strings.Cut(server, ":")
if found {
return host
}
return server
}
+145
View File
@@ -0,0 +1,145 @@
package irc
import (
"testing"
"github.com/sipeed/picoclaw/pkg/bus"
"github.com/sipeed/picoclaw/pkg/config"
)
func TestNewIRCChannel(t *testing.T) {
msgBus := bus.NewMessageBus()
t.Run("missing server", func(t *testing.T) {
cfg := config.IRCConfig{Nick: "bot"}
_, err := NewIRCChannel(cfg, msgBus)
if err == nil {
t.Error("expected error for missing server, got nil")
}
})
t.Run("missing nick", func(t *testing.T) {
cfg := config.IRCConfig{Server: "irc.example.com:6667"}
_, err := NewIRCChannel(cfg, msgBus)
if err == nil {
t.Error("expected error for missing nick, got nil")
}
})
t.Run("valid config", func(t *testing.T) {
cfg := config.IRCConfig{
Server: "irc.example.com:6667",
Nick: "testbot",
Channels: []string{"#test"},
}
ch, err := NewIRCChannel(cfg, msgBus)
if err != nil {
t.Fatalf("unexpected error: %v", err)
}
if ch.Name() != "irc" {
t.Errorf("Name() = %q, want %q", ch.Name(), "irc")
}
if ch.IsRunning() {
t.Error("new channel should not be running")
}
})
}
func TestExtractHost(t *testing.T) {
tests := []struct {
server string
want string
}{
{"irc.libera.chat:6697", "irc.libera.chat"},
{"localhost:6667", "localhost"},
{"irc.example.com", "irc.example.com"},
{"", ""},
}
for _, tt := range tests {
t.Run(tt.server, func(t *testing.T) {
got := extractHost(tt.server)
if got != tt.want {
t.Errorf("extractHost(%q) = %q, want %q", tt.server, got, tt.want)
}
})
}
}
func TestNickMentionedAt(t *testing.T) {
tests := []struct {
name string
content string
nick string
want int
}{
{"colon prefix", "bot: hello", "bot", 0},
{"comma prefix", "bot, hello", "bot", 0},
{"case insensitive", "BOT: hello", "bot", 0},
{"word boundary mid", "hey bot what's up", "bot", 4},
{"no mention", "hello world", "bot", -1},
{"substring mismatch", "robotics are cool", "bot", -1},
{"nick at end", "hello bot", "bot", 6},
{"empty content", "", "bot", -1},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
got := nickMentionedAt(tt.content, tt.nick)
if got != tt.want {
t.Errorf("nickMentionedAt(%q, %q) = %d, want %d", tt.content, tt.nick, got, tt.want)
}
})
}
}
func TestIsBotMentioned(t *testing.T) {
tests := []struct {
name string
content string
nick string
want bool
}{
{"colon prefix", "bot: hello", "bot", true},
{"comma prefix", "bot, hello", "bot", true},
{"case insensitive", "BOT: hello", "bot", true},
{"word boundary mid", "hey bot what's up", "bot", true},
{"no mention", "hello world", "bot", false},
{"substring mismatch", "robotics are cool", "bot", false},
{"nick at end", "hello bot", "bot", true},
{"empty content", "", "bot", false},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
got := isBotMentioned(tt.content, tt.nick)
if got != tt.want {
t.Errorf("isBotMentioned(%q, %q) = %v, want %v", tt.content, tt.nick, got, tt.want)
}
})
}
}
func TestStripBotMention(t *testing.T) {
tests := []struct {
name string
content string
nick string
want string
}{
{"colon prefix", "bot: hello there", "bot", "hello there"},
{"comma prefix", "bot, help me", "bot", "help me"},
{"case insensitive", "BOT: hello", "bot", "hello"},
{"no prefix match", "hello bot", "bot", "hello bot"},
{"only prefix", "bot:", "bot", ""},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
got := stripBotMention(tt.content, tt.nick)
if got != tt.want {
t.Errorf("stripBotMention(%q, %q) = %q, want %q", tt.content, tt.nick, got, tt.want)
}
})
}
}
+5
View File
@@ -62,6 +62,7 @@ var channelRateConfig = map[string]float64{
"discord": 1,
"slack": 1,
"line": 10,
"irc": 2,
}
type channelWorker struct {
@@ -269,6 +270,10 @@ func (m *Manager) initChannels() error {
m.initChannel("pico", "Pico")
}
if m.config.Channels.IRC.Enabled && m.config.Channels.IRC.Server != "" {
m.initChannel("irc", "IRC")
}
logger.InfoCF("channels", "Channel initialization completed", map[string]any{
"enabled_channels": len(m.channels),
})
+42 -8
View File
@@ -88,7 +88,7 @@ func NewTelegramChannel(cfg *config.Config, bus *bus.MessageBus) (*TelegramChann
telegramCfg,
bus,
telegramCfg.AllowFrom,
channels.WithMaxMessageLength(4096),
channels.WithMaxMessageLength(4000),
channels.WithGroupTrigger(telegramCfg.GroupTrigger),
channels.WithReasoningChannelID(telegramCfg.ReasoningChannelID),
)
@@ -173,30 +173,64 @@ func (c *TelegramChannel) Send(ctx context.Context, msg bus.OutboundMessage) err
return fmt.Errorf("invalid chat ID %s: %w", msg.ChatID, channels.ErrSendFailed)
}
htmlContent := markdownToTelegramHTML(msg.Content)
if msg.Content == "" {
return nil
}
// Typing/placeholder handled by Manager.preSend — just send the message
tgMsg := tu.Message(tu.ID(chatID), htmlContent)
// The Manager already splits messages to ≤4000 chars (WithMaxMessageLength),
// so msg.Content is guaranteed to be within that limit. We still need to
// check if HTML expansion pushes it beyond Telegram's 4096-char API limit.
queue := []string{msg.Content}
for len(queue) > 0 {
chunk := queue[0]
queue = queue[1:]
htmlContent := markdownToTelegramHTML(chunk)
if len([]rune(htmlContent)) > 4096 {
ratio := float64(len([]rune(chunk))) / float64(len([]rune(htmlContent)))
smallerLen := int(float64(4096) * ratio * 0.95) // 5% safety margin
if smallerLen < 100 {
smallerLen = 100
}
// Push sub-chunks back to the front of the queue for
// re-validation instead of sending them blindly.
subChunks := channels.SplitMessage(chunk, smallerLen)
queue = append(subChunks, queue...)
continue
}
if err := c.sendHTMLChunk(ctx, chatID, htmlContent, chunk, msg.ReplyToMessageID); err != nil {
return err
}
}
return nil
}
// sendHTMLChunk sends a single HTML message, falling back to the original
// markdown as plain text on parse failure so users never see raw HTML tags.
func (c *TelegramChannel) sendHTMLChunk(ctx context.Context, chatID int64, htmlContent, mdFallback, replyToID string) error { tgMsg := tu.Message(tu.ID(chatID), htmlContent)
tgMsg.ParseMode = telego.ModeHTML
if msg.ReplyToMessageID != "" {
if mid, parseErr := strconv.Atoi(msg.ReplyToMessageID); parseErr == nil {
if replyToID != "" {
if mid, parseErr := strconv.Atoi(replyToID); parseErr == nil {
tgMsg.ReplyParameters = &telego.ReplyParameters{
MessageID: mid,
}
}
}
if _, err = c.bot.SendMessage(ctx, tgMsg); err != nil {
if _, err := c.bot.SendMessage(ctx, tgMsg); err != nil {
logger.ErrorCF("telegram", "HTML parse failed, falling back to plain text", map[string]any{
"error": err.Error(),
})
tgMsg.Text = mdFallback
tgMsg.ParseMode = ""
if _, err = c.bot.SendMessage(ctx, tgMsg); err != nil {
return fmt.Errorf("telegram send: %w", channels.ErrTemporary)
}
}
return nil
}
+273
View File
@@ -0,0 +1,273 @@
package telegram
import (
"context"
"encoding/json"
"errors"
"strings"
"testing"
"github.com/mymmrac/telego"
ta "github.com/mymmrac/telego/telegoapi"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
"github.com/sipeed/picoclaw/pkg/bus"
"github.com/sipeed/picoclaw/pkg/channels"
)
const testToken = "1234567890:aaaabbbbaaaabbbbaaaabbbbaaaabbbbccc"
// stubCaller implements ta.Caller for testing.
type stubCaller struct {
calls []stubCall
callFn func(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error)
}
type stubCall struct {
URL string
Data *ta.RequestData
}
func (s *stubCaller) Call(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error) {
s.calls = append(s.calls, stubCall{URL: url, Data: data})
return s.callFn(ctx, url, data)
}
// stubConstructor implements ta.RequestConstructor for testing.
type stubConstructor struct{}
func (s *stubConstructor) JSONRequest(parameters any) (*ta.RequestData, error) {
return &ta.RequestData{}, nil
}
func (s *stubConstructor) MultipartRequest(
parameters map[string]string,
files map[string]ta.NamedReader,
) (*ta.RequestData, error) {
return &ta.RequestData{}, nil
}
// successResponse returns a ta.Response that telego will treat as a successful SendMessage.
func successResponse(t *testing.T) *ta.Response {
t.Helper()
msg := &telego.Message{MessageID: 1}
b, err := json.Marshal(msg)
require.NoError(t, err)
return &ta.Response{Ok: true, Result: b}
}
// newTestChannel creates a TelegramChannel with a mocked bot for unit testing.
func newTestChannel(t *testing.T, caller *stubCaller) *TelegramChannel {
t.Helper()
bot, err := telego.NewBot(testToken,
telego.WithAPICaller(caller),
telego.WithRequestConstructor(&stubConstructor{}),
telego.WithDiscardLogger(),
)
require.NoError(t, err)
base := channels.NewBaseChannel("telegram", nil, nil, nil,
channels.WithMaxMessageLength(4000),
)
base.SetRunning(true)
return &TelegramChannel{
BaseChannel: base,
bot: bot,
chatIDs: make(map[string]int64),
}
}
func TestSend_EmptyContent(t *testing.T) {
caller := &stubCaller{
callFn: func(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error) {
t.Fatal("SendMessage should not be called for empty content")
return nil, nil
},
}
ch := newTestChannel(t, caller)
err := ch.Send(context.Background(), bus.OutboundMessage{
ChatID: "12345",
Content: "",
})
assert.NoError(t, err)
assert.Empty(t, caller.calls, "no API calls should be made for empty content")
}
func TestSend_ShortMessage_SingleCall(t *testing.T) {
caller := &stubCaller{
callFn: func(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error) {
return successResponse(t), nil
},
}
ch := newTestChannel(t, caller)
err := ch.Send(context.Background(), bus.OutboundMessage{
ChatID: "12345",
Content: "Hello, world!",
})
assert.NoError(t, err)
assert.Len(t, caller.calls, 1, "short message should result in exactly one SendMessage call")
}
func TestSend_LongMessage_SingleCall(t *testing.T) {
// With WithMaxMessageLength(4000), the Manager pre-splits messages before
// they reach Send(). A message at exactly 4000 chars should go through
// as a single SendMessage call (no re-split needed since HTML expansion
// won't exceed 4096 for plain text).
caller := &stubCaller{
callFn: func(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error) {
return successResponse(t), nil
},
}
ch := newTestChannel(t, caller)
longContent := strings.Repeat("a", 4000)
err := ch.Send(context.Background(), bus.OutboundMessage{
ChatID: "12345",
Content: longContent,
})
assert.NoError(t, err)
assert.Len(t, caller.calls, 1, "pre-split message within limit should result in one SendMessage call")
}
func TestSend_HTMLFallback_PerChunk(t *testing.T) {
callCount := 0
caller := &stubCaller{
callFn: func(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error) {
callCount++
// Fail on odd calls (HTML attempt), succeed on even calls (plain text fallback)
if callCount%2 == 1 {
return nil, errors.New("Bad Request: can't parse entities")
}
return successResponse(t), nil
},
}
ch := newTestChannel(t, caller)
err := ch.Send(context.Background(), bus.OutboundMessage{
ChatID: "12345",
Content: "Hello **world**",
})
assert.NoError(t, err)
// One short message → 1 HTML attempt (fail) + 1 plain text fallback (success) = 2 calls
assert.Equal(t, 2, len(caller.calls), "should have HTML attempt + plain text fallback")
}
func TestSend_HTMLFallback_BothFail(t *testing.T) {
caller := &stubCaller{
callFn: func(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error) {
return nil, errors.New("send failed")
},
}
ch := newTestChannel(t, caller)
err := ch.Send(context.Background(), bus.OutboundMessage{
ChatID: "12345",
Content: "Hello",
})
assert.Error(t, err)
assert.True(t, errors.Is(err, channels.ErrTemporary), "error should wrap ErrTemporary")
assert.Equal(t, 2, len(caller.calls), "should have HTML attempt + plain text attempt")
}
func TestSend_LongMessage_HTMLFallback_StopsOnError(t *testing.T) {
// With a long message that gets split into 2 chunks, if both HTML and
// plain text fail on the first chunk, Send should return early.
caller := &stubCaller{
callFn: func(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error) {
return nil, errors.New("send failed")
},
}
ch := newTestChannel(t, caller)
longContent := strings.Repeat("x", 4001)
err := ch.Send(context.Background(), bus.OutboundMessage{
ChatID: "12345",
Content: longContent,
})
assert.Error(t, err)
// Should fail on the first chunk (2 calls: HTML + fallback), never reaching the second chunk.
assert.Equal(t, 2, len(caller.calls), "should stop after first chunk fails both HTML and plain text")
}
func TestSend_MarkdownShortButHTMLLong_MultipleCalls(t *testing.T) {
caller := &stubCaller{
callFn: func(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error) {
return successResponse(t), nil
},
}
ch := newTestChannel(t, caller)
// Create markdown whose length is <= 4000 but whose HTML expansion is much longer.
// "**a** " (6 chars) becomes "<b>a</b> " (9 chars) in HTML, so repeating it many times
// yields HTML that exceeds Telegram's limit while markdown stays within it.
markdownContent := strings.Repeat("**a** ", 600) // 3600 chars markdown, HTML ~5400+ chars
assert.LessOrEqual(t, len([]rune(markdownContent)), 4000, "markdown content must not exceed chunk size")
htmlExpanded := markdownToTelegramHTML(markdownContent)
assert.Greater(
t, len([]rune(htmlExpanded)), 4096,
"HTML expansion must exceed Telegram limit for this test to be meaningful",
)
err := ch.Send(context.Background(), bus.OutboundMessage{
ChatID: "12345",
Content: markdownContent,
})
assert.NoError(t, err)
assert.Greater(
t, len(caller.calls), 1,
"markdown-short but HTML-long message should be split into multiple SendMessage calls",
)
}
func TestSend_NotRunning(t *testing.T) {
caller := &stubCaller{
callFn: func(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error) {
t.Fatal("should not be called")
return nil, nil
},
}
ch := newTestChannel(t, caller)
ch.SetRunning(false)
err := ch.Send(context.Background(), bus.OutboundMessage{
ChatID: "12345",
Content: "Hello",
})
assert.ErrorIs(t, err, channels.ErrNotRunning)
assert.Empty(t, caller.calls)
}
func TestSend_InvalidChatID(t *testing.T) {
caller := &stubCaller{
callFn: func(ctx context.Context, url string, data *ta.RequestData) (*ta.Response, error) {
t.Fatal("should not be called")
return nil, nil
},
}
ch := newTestChannel(t, caller)
err := ch.Send(context.Background(), bus.OutboundMessage{
ChatID: "not-a-number",
Content: "Hello",
})
assert.Error(t, err)
assert.True(t, errors.Is(err, channels.ErrSendFailed), "error should wrap ErrSendFailed")
assert.Empty(t, caller.calls)
}
+20
View File
@@ -232,6 +232,7 @@ type ChannelsConfig struct {
WeComApp WeComAppConfig `json:"wecom_app"`
WeComAIBot WeComAIBotConfig `json:"wecom_aibot"`
Pico PicoConfig `json:"pico"`
IRC IRCConfig `json:"irc"`
}
// GroupTriggerConfig controls when the bot responds in group chats.
@@ -415,6 +416,25 @@ type PicoConfig struct {
Placeholder PlaceholderConfig `json:"placeholder,omitempty"`
}
type IRCConfig struct {
Enabled bool `json:"enabled" env:"PICOCLAW_CHANNELS_IRC_ENABLED"`
Server string `json:"server" env:"PICOCLAW_CHANNELS_IRC_SERVER"`
TLS bool `json:"tls" env:"PICOCLAW_CHANNELS_IRC_TLS"`
Nick string `json:"nick" env:"PICOCLAW_CHANNELS_IRC_NICK"`
User string `json:"user,omitempty" env:"PICOCLAW_CHANNELS_IRC_USER"`
RealName string `json:"real_name,omitempty" env:"PICOCLAW_CHANNELS_IRC_REAL_NAME"`
Password string `json:"password" env:"PICOCLAW_CHANNELS_IRC_PASSWORD"`
NickServPassword string `json:"nickserv_password" env:"PICOCLAW_CHANNELS_IRC_NICKSERV_PASSWORD"`
SASLUser string `json:"sasl_user" env:"PICOCLAW_CHANNELS_IRC_SASL_USER"`
SASLPassword string `json:"sasl_password" env:"PICOCLAW_CHANNELS_IRC_SASL_PASSWORD"`
Channels FlexibleStringSlice `json:"channels" env:"PICOCLAW_CHANNELS_IRC_CHANNELS"`
RequestCaps FlexibleStringSlice `json:"request_caps,omitempty" env:"PICOCLAW_CHANNELS_IRC_REQUEST_CAPS"`
AllowFrom FlexibleStringSlice `json:"allow_from" env:"PICOCLAW_CHANNELS_IRC_ALLOW_FROM"`
GroupTrigger GroupTriggerConfig `json:"group_trigger,omitempty"`
Typing TypingConfig `json:"typing,omitempty"`
ReasoningChannelID string `json:"reasoning_channel_id" env:"PICOCLAW_CHANNELS_IRC_REASONING_CHANNEL_ID"`
}
type HeartbeatConfig struct {
Enabled bool `json:"enabled" env:"PICOCLAW_HEARTBEAT_ENABLED"`
Interval int `json:"interval" env:"PICOCLAW_HEARTBEAT_INTERVAL"` // minutes, min 5
+2 -2
View File
@@ -8,7 +8,7 @@ import (
func TestSpawnTool_Execute_EmptyTask(t *testing.T) {
provider := &MockLLMProvider{}
manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
tool := NewSpawnTool(manager)
ctx := context.Background()
@@ -42,7 +42,7 @@ func TestSpawnTool_Execute_EmptyTask(t *testing.T) {
func TestSpawnTool_Execute_ValidTask(t *testing.T) {
provider := &MockLLMProvider{}
manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
tool := NewSpawnTool(manager)
ctx := context.Background()
-18
View File
@@ -6,7 +6,6 @@ import (
"sync"
"time"
"github.com/sipeed/picoclaw/pkg/bus"
"github.com/sipeed/picoclaw/pkg/providers"
)
@@ -27,7 +26,6 @@ type SubagentManager struct {
mu sync.RWMutex
provider providers.LLMProvider
defaultModel string
bus *bus.MessageBus
workspace string
tools *ToolRegistry
maxIterations int
@@ -41,13 +39,11 @@ type SubagentManager struct {
func NewSubagentManager(
provider providers.LLMProvider,
defaultModel, workspace string,
bus *bus.MessageBus,
) *SubagentManager {
return &SubagentManager{
tasks: make(map[string]*SubagentTask),
provider: provider,
defaultModel: defaultModel,
bus: bus,
workspace: workspace,
tools: NewToolRegistry(),
maxIterations: 10,
@@ -214,20 +210,6 @@ After completing the task, provide a clear summary of what was done.`
Async: false,
}
}
// Send announce message back to main agent
if sm.bus != nil {
announceContent := fmt.Sprintf("Task '%s' completed.\n\nResult:\n%s", task.Label, task.Result)
pubCtx, pubCancel := context.WithTimeout(context.Background(), 5*time.Second)
defer pubCancel()
sm.bus.PublishInbound(pubCtx, bus.InboundMessage{
Channel: "system",
SenderID: fmt.Sprintf("subagent:%s", task.ID),
// Format: "original_channel:original_chat_id" for routing back
ChatID: fmt.Sprintf("%s:%s", task.OriginChannel, task.OriginChatID),
Content: announceContent,
})
}
}
func (sm *SubagentManager) GetTask(taskID string) (*SubagentTask, bool) {
+9 -14
View File
@@ -5,7 +5,6 @@ import (
"strings"
"testing"
"github.com/sipeed/picoclaw/pkg/bus"
"github.com/sipeed/picoclaw/pkg/providers"
)
@@ -47,7 +46,7 @@ func (m *MockLLMProvider) GetContextWindow() int {
func TestSubagentManager_SetLLMOptions_AppliesToRunToolLoop(t *testing.T) {
provider := &MockLLMProvider{}
manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
manager.SetLLMOptions(2048, 0.6)
tool := NewSubagentTool(manager)
@@ -73,7 +72,7 @@ func TestSubagentManager_SetLLMOptions_AppliesToRunToolLoop(t *testing.T) {
// TestSubagentTool_Name verifies tool name
func TestSubagentTool_Name(t *testing.T) {
provider := &MockLLMProvider{}
manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
tool := NewSubagentTool(manager)
if tool.Name() != "subagent" {
@@ -84,7 +83,7 @@ func TestSubagentTool_Name(t *testing.T) {
// TestSubagentTool_Description verifies tool description
func TestSubagentTool_Description(t *testing.T) {
provider := &MockLLMProvider{}
manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
tool := NewSubagentTool(manager)
desc := tool.Description()
@@ -99,7 +98,7 @@ func TestSubagentTool_Description(t *testing.T) {
// TestSubagentTool_Parameters verifies tool parameters schema
func TestSubagentTool_Parameters(t *testing.T) {
provider := &MockLLMProvider{}
manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
tool := NewSubagentTool(manager)
params := tool.Parameters()
@@ -149,8 +148,7 @@ func TestSubagentTool_Parameters(t *testing.T) {
// TestSubagentTool_Execute_Success tests successful execution
func TestSubagentTool_Execute_Success(t *testing.T) {
provider := &MockLLMProvider{}
msgBus := bus.NewMessageBus()
manager := NewSubagentManager(provider, "test-model", "/tmp/test", msgBus)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
tool := NewSubagentTool(manager)
ctx := WithToolContext(context.Background(), "telegram", "chat-123")
@@ -204,8 +202,7 @@ func TestSubagentTool_Execute_Success(t *testing.T) {
// TestSubagentTool_Execute_NoLabel tests execution without label
func TestSubagentTool_Execute_NoLabel(t *testing.T) {
provider := &MockLLMProvider{}
msgBus := bus.NewMessageBus()
manager := NewSubagentManager(provider, "test-model", "/tmp/test", msgBus)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
tool := NewSubagentTool(manager)
ctx := context.Background()
@@ -228,7 +225,7 @@ func TestSubagentTool_Execute_NoLabel(t *testing.T) {
// TestSubagentTool_Execute_MissingTask tests error handling for missing task
func TestSubagentTool_Execute_MissingTask(t *testing.T) {
provider := &MockLLMProvider{}
manager := NewSubagentManager(provider, "test-model", "/tmp/test", nil)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
tool := NewSubagentTool(manager)
ctx := context.Background()
@@ -278,8 +275,7 @@ func TestSubagentTool_Execute_NilManager(t *testing.T) {
// TestSubagentTool_Execute_ContextPassing verifies context is properly used
func TestSubagentTool_Execute_ContextPassing(t *testing.T) {
provider := &MockLLMProvider{}
msgBus := bus.NewMessageBus()
manager := NewSubagentManager(provider, "test-model", "/tmp/test", msgBus)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
tool := NewSubagentTool(manager)
channel := "test-channel"
@@ -304,8 +300,7 @@ func TestSubagentTool_Execute_ContextPassing(t *testing.T) {
func TestSubagentTool_ForUserTruncation(t *testing.T) {
// Create a mock provider that returns very long content
provider := &MockLLMProvider{}
msgBus := bus.NewMessageBus()
manager := NewSubagentManager(provider, "test-model", "/tmp/test", msgBus)
manager := NewSubagentManager(provider, "test-model", "/tmp/test")
tool := NewSubagentTool(manager)
ctx := context.Background()
+56
View File
@@ -0,0 +1,56 @@
#!/bin/sh
# Starts a local Ergo IRC server for testing the IRC channel.
#
# Requirements: docker
# Usage: ./scripts/test-irc.sh
set -e
CONTAINER_NAME="picoclaw-test-ergo"
IRC_PORT=6667
# Clean up any previous instance
docker rm -f "$CONTAINER_NAME" >/dev/null 2>&1 || true
echo "Starting Ergo IRC server on port $IRC_PORT..."
docker run -d \
--name "$CONTAINER_NAME" \
-p "$IRC_PORT:6667" \
ghcr.io/ergochat/ergo:stable
for i in $(seq 1 10); do
if nc -z localhost "$IRC_PORT" 2>/dev/null; then
break
fi
if [ "$i" -eq 10 ]; then
echo "ERROR: Server did not start within 10s"
exit 1
fi
sleep 1
done
echo ""
echo "IRC server ready on localhost:$IRC_PORT"
echo ""
echo "Add this to your ~/.picoclaw/config.json under \"channels\":"
echo ""
echo ' "irc": {'
echo ' "enabled": true,'
echo ' "server": "localhost:6667",'
echo ' "tls": false,'
echo ' "nick": "picobot",'
echo ' "channels": ["#test"],'
echo ' "allow_from": [],'
echo ' "group_trigger": { "mention_only": true }'
echo ' }'
echo ""
echo "Then run picoclaw:"
echo " cd packages/picoclaw && go run ./cmd/picoclaw gateway"
echo ""
echo "Connect with an IRC client:"
echo " irssi: /connect localhost $IRC_PORT"
echo " weechat: /server add test localhost/$IRC_PORT && /connect test"
echo " Join #test, then: picobot: hello"
echo ""
echo "To stop the IRC server:"
echo " docker rm -f $CONTAINER_NAME"