lzh/picoclaw
1
0
Fork 0
mirror of https://github.com/sipeed/picoclaw.git synced 2026-06-12 18:08:54 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
329322075df3f077a39530b9cd780fc2dd2b8396
picoclaw/pkg/config/defaults.go
T
Amir Mamaghani 71134babb9 feat(telegram): stream LLM responses via sendMessageDraft (#1101) 
* feat(telegram): stream LLM responses in real-time via sendMessageDraft

Implements real-time token streaming to Telegram using the sendMessageDraft
API (telego v1.6.0). Instead of showing only a "Thinking..." placeholder
until the full response arrives, users now see partial LLM output appear
in the chat as it's generated.

The streaming pipeline threads through all layers:

- StreamingProvider interface (providers/types.go): opt-in ChatStream()
  method that receives an onChunk callback with accumulated text
- OpenAI-compatible SSE streaming (openai_compat/provider.go): parses
  SSE events with stream:true, handles text deltas and tool call assembly
- Anthropic native streaming (anthropic/provider.go): uses SDK's
  NewStreaming() for direct Anthropic API connections
- HTTPProvider delegation (http_provider.go): delegates ChatStream to
  the underlying openai_compat provider
- StreamingCapable + Streamer interfaces (channels/interfaces.go):
  opt-in channel capability like TypingCapable/PlaceholderCapable
- Telegram streamer (telegram/telegram.go): BeginStream returns a
  telegramStreamer that throttles sendMessageDraft calls (3s/200 chars)
  with graceful degradation on API errors
- StreamDelegate bridge (bus/bus.go): decouples agent loop from channel
  manager without tight imports
- Manager integration (manager.go): implements StreamDelegate, tracks
  streamActive state, coordinates with placeholder editing
- Agent loop (loop.go): uses ChatStream when both provider and channel
  support streaming, cancels stream on tool calls, skips PublishOutbound
  when Finalize already delivered the message

Graceful degradation:
- Bots without forum/topics mode: first sendMessageDraft error sets
  failed=true, subsequent Updates become no-ops, Finalize still delivers
  via SendMessage. User sees normal non-streaming behavior.
- Non-streaming providers: type assertion fails, falls back to Chat()
- Config opt-out: streaming.enabled (default true) in telegram config

Closes #1098

* fix(telegram): delete placeholder message when streaming delivers response

When streaming was active, the "Thinking..." placeholder message stayed
in the chat because preSend only deleted the tracking entry without
removing the actual Telegram message. Now preSend deletes the placeholder
via the new MessageDeleter interface when streamActive is set.

* refactor(streaming): remove dead code and simplify streaming wiring

- Delete unused Anthropic ChatStream/parseStream (-131 lines) — factory
  creates HTTPProvider for all OpenAI-compat providers including OpenRouter
- Simplify runLLMIteration from 4 to 3 return values (remove unused
  streamed bool)
- Replace managerStreamer struct with finalizeHookStreamer using embedding
  (Update/Cancel promoted, only Finalize overridden)

* fix(streaming): skip streamer acquisition when SendResponse is false

Heartbeat messages set SendResponse=false but the streaming path
was unconditionally acquiring a streamer, causing HEARTBEAT_OK to
leak to Telegram via streamer.Finalize().

* fix(streaming): guard streamer for non-sendable messages, add streaming config

Skip streamer acquisition for heartbeat (NoHistory=true), preventing
HEARTBEAT_OK from leaking to Telegram via streamer.Finalize().

Add streaming.enabled to Telegram defaults and example config.

* feat(telegram): stream LLM responses in real-time via sendMessageDraft

Implements real-time token streaming to Telegram using the sendMessageDraft
API (telego v1.6.0). Instead of showing only a "Thinking..." placeholder
until the full response arrives, users now see partial LLM output appear
in the chat as it's generated.

The streaming pipeline threads through all layers:

- StreamingProvider interface (providers/types.go): opt-in ChatStream()
  method that receives an onChunk callback with accumulated text
- OpenAI-compatible SSE streaming (openai_compat/provider.go): parses
  SSE events with stream:true, handles text deltas and tool call assembly
- Anthropic native streaming (anthropic/provider.go): uses SDK's
  NewStreaming() for direct Anthropic API connections
- HTTPProvider delegation (http_provider.go): delegates ChatStream to
  the underlying openai_compat provider
- StreamingCapable + Streamer interfaces (channels/interfaces.go):
  opt-in channel capability like TypingCapable/PlaceholderCapable
- Telegram streamer (telegram/telegram.go): BeginStream returns a
  telegramStreamer that throttles sendMessageDraft calls (3s/200 chars)
  with graceful degradation on API errors
- StreamDelegate bridge (bus/bus.go): decouples agent loop from channel
  manager without tight imports
- Manager integration (manager.go): implements StreamDelegate, tracks
  streamActive state, coordinates with placeholder editing
- Agent loop (loop.go): uses ChatStream when both provider and channel
  support streaming, cancels stream on tool calls, skips PublishOutbound
  when Finalize already delivered the message

Graceful degradation:
- Bots without forum/topics mode: first sendMessageDraft error sets
  failed=true, subsequent Updates become no-ops, Finalize still delivers
  via SendMessage. User sees normal non-streaming behavior.
- Non-streaming providers: type assertion fails, falls back to Chat()
- Config opt-out: streaming.enabled (default true) in telegram config

Closes #1098

* fix(telegram): delete placeholder message when streaming delivers response

When streaming was active, the "Thinking..." placeholder message stayed
in the chat because preSend only deleted the tracking entry without
removing the actual Telegram message. Now preSend deletes the placeholder
via the new MessageDeleter interface when streamActive is set.

* refactor(streaming): remove dead code and simplify streaming wiring

- Delete unused Anthropic ChatStream/parseStream (-131 lines) — factory
  creates HTTPProvider for all OpenAI-compat providers including OpenRouter
- Simplify runLLMIteration from 4 to 3 return values (remove unused
  streamed bool)
- Replace managerStreamer struct with finalizeHookStreamer using embedding
  (Update/Cancel promoted, only Finalize overridden)

* fix(streaming): skip streamer acquisition when SendResponse is false

Heartbeat messages set SendResponse=false but the streaming path
was unconditionally acquiring a streamer, causing HEARTBEAT_OK to
leak to Telegram via streamer.Finalize().

* fix(streaming): guard streamer for non-sendable messages, add streaming config

Skip streamer acquisition for heartbeat (NoHistory=true), preventing
HEARTBEAT_OK from leaking to Telegram via streamer.Finalize().

Add streaming.enabled to Telegram defaults and example config.

* fix(picoclaw): add missing closing brace for StreamingProvider interface

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix: resolve golangci-lint formatting issues

Fix gci import ordering in telegram and anthropic provider, and break
long function signature in openai_compat provider to satisfy golines.

* fix: address code review feedback on streaming PR

- Deduplicate Streamer interface: alias channels.Streamer to bus.Streamer
  to prevent type drift across packages
- Increase SSE scanner buffer to 10MB max to handle large single-line
  responses that exceed bufio.Scanner's 64KB default
- Switch draftID generation from math/rand to crypto/rand for
  collision-resistant random IDs
- Add context cancellation check in SSE parsing loop so cancelled
  streams stop processing immediately
- Log Finalize failures with chat_id and content length for debugging
  silent message delivery failures

* feat: make streaming throttle interval and min growth configurable

Move hardcoded streamThrottleInterval (3s) and streamMinGrowth (200)
into StreamingConfig so they can be tuned per deployment via config
or environment variables.

* fix(telegram): use parseTelegramChatID in DeleteMessage and BeginStream

These two functions called undefined parseChatID. Use
parseTelegramChatID with _ for the unused threadID instead of adding
a wrapper function. Fixes all three CI checks.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* fix(streaming): set streamActive only after successful Finalize

Move onFinalize hook to run after Streamer.Finalize succeeds, so that
if Finalize fails the streamActive flag stays false and the regular
placeholder fallback path remains available.

Addresses review feedback from @alexhoshina.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-20 21:04:14 +08:00

571 lines
14 KiB
Go
Raw Blame History

// PicoClaw - Ultra-lightweight personal AI agent
// License: MIT
//
// Copyright (c) 2026 PicoClaw contributors
package config
import (
"os"
"path/filepath"
)
// DefaultConfig returns the default configuration for PicoClaw.
func DefaultConfig() *Config {
// Determine the base path for the workspace.
// Priority: $PICOCLAW_HOME > ~/.picoclaw
var homePath string
if picoclawHome := os.Getenv(EnvHome); picoclawHome != "" {
homePath = picoclawHome
} else {
userHome, _ := os.UserHomeDir()
homePath = filepath.Join(userHome, ".picoclaw")
}
workspacePath := filepath.Join(homePath, "workspace")
return &Config{
Agents: AgentsConfig{
Defaults: AgentDefaults{
Workspace: workspacePath,
RestrictToWorkspace: true,
Provider: "",
Model: "",
MaxTokens: 32768,
Temperature: nil, // nil means use provider default
MaxToolIterations: 50,
SummarizeMessageThreshold: 20,
SummarizeTokenPercent: 75,
ToolFeedback: ToolFeedbackConfig{
Enabled: true,
MaxArgsLength: 300,
},
},
},
Bindings: []AgentBinding{},
Session: SessionConfig{
DMScope: "per-channel-peer",
},
Channels: ChannelsConfig{
WhatsApp: WhatsAppConfig{
Enabled: false,
BridgeURL: "ws://localhost:3001",
UseNative: false,
SessionStorePath: "",
AllowFrom: FlexibleStringSlice{},
},
Telegram: TelegramConfig{
Enabled: false,
Token: "",
AllowFrom: FlexibleStringSlice{},
Typing: TypingConfig{Enabled: true},
Placeholder: PlaceholderConfig{
Enabled: true,
Text: "Thinking... 💭",
},
Streaming: StreamingConfig{Enabled: true, ThrottleSeconds: 3, MinGrowthChars: 200},
UseMarkdownV2: false,
},
Feishu: FeishuConfig{
Enabled: false,
AppID: "",
AppSecret: "",
EncryptKey: "",
VerificationToken: "",
AllowFrom: FlexibleStringSlice{},
},
Discord: DiscordConfig{
Enabled: false,
Token: "",
AllowFrom: FlexibleStringSlice{},
MentionOnly: false,
},
MaixCam: MaixCamConfig{
Enabled: false,
Host: "0.0.0.0",
Port: 18790,
AllowFrom: FlexibleStringSlice{},
},
QQ: QQConfig{
Enabled: false,
AppID: "",
AppSecret: "",
AllowFrom: FlexibleStringSlice{},
MaxMessageLength: 2000,
MaxBase64FileSizeMiB: 0,
},
DingTalk: DingTalkConfig{
Enabled: false,
ClientID: "",
ClientSecret: "",
AllowFrom: FlexibleStringSlice{},
},
Slack: SlackConfig{
Enabled: false,
BotToken: "",
AppToken: "",
AllowFrom: FlexibleStringSlice{},
},
Matrix: MatrixConfig{
Enabled: false,
Homeserver: "https://matrix.org",
UserID: "",
AccessToken: "",
DeviceID: "",
JoinOnInvite: true,
AllowFrom: FlexibleStringSlice{},
GroupTrigger: GroupTriggerConfig{
MentionOnly: true,
},
Placeholder: PlaceholderConfig{
Enabled: true,
Text: "Thinking... 💭",
},
},
LINE: LINEConfig{
Enabled: false,
ChannelSecret: "",
ChannelAccessToken: "",
WebhookHost: "0.0.0.0",
WebhookPort: 18791,
WebhookPath: "/webhook/line",
AllowFrom: FlexibleStringSlice{},
GroupTrigger: GroupTriggerConfig{MentionOnly: true},
},
OneBot: OneBotConfig{
Enabled: false,
WSUrl: "ws://127.0.0.1:3001",
AccessToken: "",
ReconnectInterval: 5,
GroupTriggerPrefix: []string{},
AllowFrom: FlexibleStringSlice{},
},
WeCom: WeComConfig{
Enabled: false,
Token: "",
EncodingAESKey: "",
WebhookURL: "",
WebhookHost: "0.0.0.0",
WebhookPort: 18793,
WebhookPath: "/webhook/wecom",
AllowFrom: FlexibleStringSlice{},
ReplyTimeout: 5,
},
WeComApp: WeComAppConfig{
Enabled: false,
CorpID: "",
CorpSecret: "",
AgentID: 0,
Token: "",
EncodingAESKey: "",
WebhookHost: "0.0.0.0",
WebhookPort: 18792,
WebhookPath: "/webhook/wecom-app",
AllowFrom: FlexibleStringSlice{},
ReplyTimeout: 5,
},
WeComAIBot: WeComAIBotConfig{
Enabled: false,
Token: "",
EncodingAESKey: "",
WebhookPath: "/webhook/wecom-aibot",
AllowFrom: FlexibleStringSlice{},
ReplyTimeout: 5,
MaxSteps: 10,
WelcomeMessage: "Hello! I'm your AI assistant. How can I help you today?",
ProcessingMessage: DefaultWeComAIBotProcessingMessage,
},
Pico: PicoConfig{
Enabled: false,
Token: "",
PingInterval: 30,
ReadTimeout: 60,
WriteTimeout: 10,
MaxConnections: 100,
AllowFrom: FlexibleStringSlice{},
},
},
Providers: ProvidersConfig{
OpenAI: OpenAIProviderConfig{WebSearch: true},
},
ModelList: []ModelConfig{
// ============================================
// Add your API key to the model you want to use
// ============================================
// Zhipu AI (智谱) - https://open.bigmodel.cn/usercenter/apikeys
{
ModelName: "glm-4.7",
Model: "zhipu/glm-4.7",
APIBase: "https://open.bigmodel.cn/api/paas/v4",
APIKey: "",
},
// OpenAI - https://platform.openai.com/api-keys
{
ModelName: "gpt-5.4",
Model: "openai/gpt-5.4",
APIBase: "https://api.openai.com/v1",
APIKey: "",
},
// Anthropic Claude - https://console.anthropic.com/settings/keys
{
ModelName: "claude-sonnet-4.6",
Model: "anthropic/claude-sonnet-4.6",
APIBase: "https://api.anthropic.com/v1",
APIKey: "",
},
// DeepSeek - https://platform.deepseek.com/
{
ModelName: "deepseek-chat",
Model: "deepseek/deepseek-chat",
APIBase: "https://api.deepseek.com/v1",
APIKey: "",
},
// Google Gemini - https://ai.google.dev/
{
ModelName: "gemini-2.0-flash",
Model: "gemini/gemini-2.0-flash-exp",
APIBase: "https://generativelanguage.googleapis.com/v1beta",
APIKey: "",
},
// Qwen (通义千问) - https://dashscope.console.aliyun.com/apiKey
{
ModelName: "qwen-plus",
Model: "qwen/qwen-plus",
APIBase: "https://dashscope.aliyuncs.com/compatible-mode/v1",
APIKey: "",
},
// Moonshot (月之暗面) - https://platform.moonshot.cn/console/api-keys
{
ModelName: "moonshot-v1-8k",
Model: "moonshot/moonshot-v1-8k",
APIBase: "https://api.moonshot.cn/v1",
APIKey: "",
},
// Groq - https://console.groq.com/keys
{
ModelName: "llama-3.3-70b",
Model: "groq/llama-3.3-70b-versatile",
APIBase: "https://api.groq.com/openai/v1",
APIKey: "",
},
// OpenRouter (100+ models) - https://openrouter.ai/keys
{
ModelName: "openrouter-auto",
Model: "openrouter/auto",
APIBase: "https://openrouter.ai/api/v1",
APIKey: "",
},
{
ModelName: "openrouter-gpt-5.4",
Model: "openrouter/openai/gpt-5.4",
APIBase: "https://openrouter.ai/api/v1",
APIKey: "",
},
// NVIDIA - https://build.nvidia.com/
{
ModelName: "nemotron-4-340b",
Model: "nvidia/nemotron-4-340b-instruct",
APIBase: "https://integrate.api.nvidia.com/v1",
APIKey: "",
},
// Cerebras - https://inference.cerebras.ai/
{
ModelName: "cerebras-llama-3.3-70b",
Model: "cerebras/llama-3.3-70b",
APIBase: "https://api.cerebras.ai/v1",
APIKey: "",
},
// Vivgrid - https://vivgrid.com
{
ModelName: "vivgrid-auto",
Model: "vivgrid/auto",
APIBase: "https://api.vivgrid.com/v1",
APIKey: "",
},
// Volcengine (火山引擎) - https://console.volcengine.com/ark
{
ModelName: "ark-code-latest",
Model: "volcengine/ark-code-latest",
APIBase: "https://ark.cn-beijing.volces.com/api/v3",
APIKey: "",
},
{
ModelName: "doubao-pro",
Model: "volcengine/doubao-pro-32k",
APIBase: "https://ark.cn-beijing.volces.com/api/v3",
APIKey: "",
},
// ShengsuanYun (神算云)
{
ModelName: "deepseek-v3",
Model: "shengsuanyun/deepseek-v3",
APIBase: "https://api.shengsuanyun.com/v1",
APIKey: "",
},
// Antigravity (Google Cloud Code Assist) - OAuth only
{
ModelName: "gemini-flash",
Model: "antigravity/gemini-3-flash",
AuthMethod: "oauth",
},
// GitHub Copilot - https://github.com/settings/tokens
{
ModelName: "copilot-gpt-5.4",
Model: "github-copilot/gpt-5.4",
APIBase: "http://localhost:4321",
AuthMethod: "oauth",
},
// Ollama (local) - https://ollama.com
{
ModelName: "llama3",
Model: "ollama/llama3",
APIBase: "http://localhost:11434/v1",
APIKey: "ollama",
},
// Mistral AI - https://console.mistral.ai/api-keys
{
ModelName: "mistral-small",
Model: "mistral/mistral-small-latest",
APIBase: "https://api.mistral.ai/v1",
APIKey: "",
},
// Avian - https://avian.io
{
ModelName: "deepseek-v3.2",
Model: "avian/deepseek/deepseek-v3.2",
APIBase: "https://api.avian.io/v1",
APIKey: "",
},
{
ModelName: "kimi-k2.5",
Model: "avian/moonshotai/kimi-k2.5",
APIBase: "https://api.avian.io/v1",
APIKey: "",
},
// Minimax - https://api.minimaxi.com/
{
ModelName: "MiniMax-M2.5",
Model: "minimax/MiniMax-M2.5",
APIBase: "https://api.minimaxi.com/v1",
APIKey: "",
},
// LongCat - https://longcat.chat/platform
{
ModelName: "LongCat-Flash-Thinking",
Model: "longcat/LongCat-Flash-Thinking",
APIBase: "https://api.longcat.chat/openai",
APIKey: "",
},
// ModelScope (魔搭社区) - https://modelscope.cn/my/tokens
{
ModelName: "modelscope-qwen",
Model: "modelscope/Qwen/Qwen3-235B-A22B-Instruct-2507",
APIBase: "https://api-inference.modelscope.cn/v1",
APIKey: "",
},
// VLLM (local) - http://localhost:8000
{
ModelName: "local-model",
Model: "vllm/custom-model",
APIBase: "http://localhost:8000/v1",
APIKey: "",
},
// Azure OpenAI - https://portal.azure.com
// model_name is a user-friendly alias; the model field's path after "azure/" is your deployment name
{
ModelName: "azure-gpt5",
Model: "azure/my-gpt5-deployment",
APIBase: "https://your-resource.openai.azure.com",
APIKey: "",
},
},
Gateway: GatewayConfig{
Host: "127.0.0.1",
Port: 18790,
HotReload: false,
},
Tools: ToolsConfig{
MediaCleanup: MediaCleanupConfig{
ToolConfig: ToolConfig{
Enabled: true,
},
MaxAge: 30,
Interval: 5,
},
Web: WebToolsConfig{
ToolConfig: ToolConfig{
Enabled: true,
},
PreferNative: true,
Proxy: "",
FetchLimitBytes: 10 * 1024 * 1024, // 10MB by default
Format: "plaintext",
Brave: BraveConfig{
Enabled: false,
APIKey: "",
APIKeys: nil,
MaxResults: 5,
},
Tavily: TavilyConfig{
Enabled: false,
APIKey: "",
APIKeys: nil,
MaxResults: 5,
},
DuckDuckGo: DuckDuckGoConfig{
Enabled: true,
MaxResults: 5,
},
Perplexity: PerplexityConfig{
Enabled: false,
APIKey: "",
APIKeys: nil,
MaxResults: 5,
},
SearXNG: SearXNGConfig{
Enabled: false,
BaseURL: "",
MaxResults: 5,
},
GLMSearch: GLMSearchConfig{
Enabled: false,
APIKey: "",
BaseURL: "https://open.bigmodel.cn/api/paas/v4/web_search",
SearchEngine: "search_std",
MaxResults: 5,
},
},
Cron: CronToolsConfig{
ToolConfig: ToolConfig{
Enabled: true,
},
ExecTimeoutMinutes: 5,
AllowCommand: true,
},
Exec: ExecConfig{
ToolConfig: ToolConfig{
Enabled: true,
},
EnableDenyPatterns: true,
AllowRemote: true,
TimeoutSeconds: 60,
},
Skills: SkillsToolsConfig{
ToolConfig: ToolConfig{
Enabled: true,
},
Registries: SkillsRegistriesConfig{
ClawHub: ClawHubRegistryConfig{
Enabled: true,
BaseURL: "https://clawhub.ai",
},
},
MaxConcurrentSearches: 2,
SearchCache: SearchCacheConfig{
MaxSize: 50,
TTLSeconds: 300,
},
},
SendFile: ToolConfig{
Enabled: true,
},
MCP: MCPConfig{
ToolConfig: ToolConfig{
Enabled: false,
},
Discovery: ToolDiscoveryConfig{
Enabled: false,
TTL: 5,
MaxSearchResults: 5,
UseBM25: true,
UseRegex: false,
},
Servers: map[string]MCPServerConfig{},
},
AppendFile: ToolConfig{
Enabled: true,
},
EditFile: ToolConfig{
Enabled: true,
},
FindSkills: ToolConfig{
Enabled: true,
},
I2C: ToolConfig{
Enabled: false, // Hardware tool - Linux only
},
InstallSkill: ToolConfig{
Enabled: true,
},
ListDir: ToolConfig{
Enabled: true,
},
Message: ToolConfig{
Enabled: true,
},
ReadFile: ReadFileToolConfig{
Enabled: true,
MaxReadFileSize: 64 * 1024, // 64KB
},
Spawn: ToolConfig{
Enabled: true,
},
SpawnStatus: ToolConfig{
Enabled: false,
},
SPI: ToolConfig{
Enabled: false, // Hardware tool - Linux only
},
Subagent: ToolConfig{
Enabled: true,
},
WebFetch: ToolConfig{
Enabled: true,
},
WriteFile: ToolConfig{
Enabled: true,
},
},
Heartbeat: HeartbeatConfig{
Enabled: true,
Interval: 30,
},
Devices: DevicesConfig{
Enabled: false,
MonitorUSB: true,
},
Voice: VoiceConfig{
EchoTranscription: false,
},
BuildInfo: BuildInfo{
Version: Version,
GitCommit: GitCommit,
BuildTime: BuildTime,
GoVersion: GoVersion,
},
}
}
Reference in New Issue View Git Blame Copy Permalink