Config Schema

Triggerfish triggerfish.yaml द्वारे configured आहे, triggerfish dive run केल्यानंतर ~/.triggerfish/triggerfish.yaml वर located. हे page प्रत्येक configuration section document करतो.

Secret References या file मधील कोणतेही string value OS keychain मध्ये

stored credential reference करण्यासाठी secret: prefix वापरू शकते. उदाहरणार्थ, apiKey: "secret:provider:anthropic:apiKey" startup वर keychain मधून value resolve करतो. Details साठी Secrets Management पहा. :::

Full Annotated Example

# =============================================================================
# triggerfish.yaml -- Complete configuration reference
# =============================================================================

# ---------------------------------------------------------------------------
# Models: LLM provider configuration and failover
# ---------------------------------------------------------------------------
models:
  # Agent completions साठी वापरला जाणारा primary model
  primary:
    provider: anthropic
    model: claude-sonnet-4-5

  # Optional: image description साठी separate vision model
  # Primary model vision support करत नसल्यास, images primary ला पोहोचण्यापूर्वी
  # automatically या model द्वारे described होतात.
  # vision: glm-4.5v

  # Streaming responses (default: true)
  # streaming: true

  # Provider-specific configuration
  # API keys secret: syntax द्वारे referenced आहेत आणि OS keychain मधून resolved होतात.
  # Set up करण्यासाठी `triggerfish dive` किंवा `triggerfish config migrate-secrets` run करा.
  providers:
    anthropic:
      model: claude-sonnet-4-5
      # apiKey: "secret:provider:anthropic:apiKey"

    openai:
      model: gpt-4o

    google:
      model: gemini-pro

    ollama:
      model: llama3
      endpoint: "http://localhost:11434"

    lmstudio:
      model: lmstudio-community/Meta-Llama-3.1-8B-Instruct-GGUF
      endpoint: "http://localhost:1234"

    openrouter:
      model: anthropic/claude-sonnet-4-5

    zenmux:
      model: openai/gpt-5

    zai:
      model: glm-4.7

  # Ordered failover chain -- primary fail झाल्यावर sequence मध्ये tried
  failover:
    - claude-haiku-4-5 # First fallback
    - gpt-4o # Second fallback
    - ollama/llama3 # Local fallback (internet आवश्यक नाही)

  # Failover behavior
  failover_config:
    max_retries: 3 # Provider बदलण्यापूर्वी per provider retries
    retry_delay_ms: 1000 # Retries दरम्यान delay
    conditions: # काय failover trigger करते
      - rate_limited # Provider returned 429
      - server_error # Provider returned 5xx
      - timeout # Request exceeded timeout

# ---------------------------------------------------------------------------
# Logging: Structured log output
# ---------------------------------------------------------------------------
logging:
  level: normal # quiet | normal | verbose | debug

# ---------------------------------------------------------------------------
# Channels: Messaging platform connections
# ---------------------------------------------------------------------------
# Secrets (bot tokens, API keys, passwords) OS keychain मध्ये stored आहेत.
# Securely enter करण्यासाठी `triggerfish config add-channel <name>` run करा.
# फक्त non-secret configuration येथे appears होते.
channels:
  telegram:
    ownerId: 123456789 # तुमचा Telegram numeric user ID
    classification: INTERNAL # Default: INTERNAL

  signal:
    endpoint: "tcp://127.0.0.1:7583" # signal-cli daemon endpoint
    account: "+14155552671" # तुमचा Signal phone number (E.164)
    classification: PUBLIC # Default: PUBLIC
    defaultGroupMode: mentioned-only # always | mentioned-only | owner-only
    groups:
      "group-id-here":
        mode: always
        classification: INTERNAL

  slack:
    classification: PUBLIC # Default: PUBLIC

  discord:
    ownerId: "your-discord-user-id" # तुमचा Discord user ID
    classification: PUBLIC # Default: PUBLIC

  whatsapp:
    phoneNumberId: "your-phone-number-id" # Meta Business Dashboard मधून
    classification: PUBLIC # Default: PUBLIC

  webchat:
    port: 8765 # Web client साठी WebSocket port
    classification: PUBLIC # Default: PUBLIC (visitors)

  email:
    smtpApiUrl: "https://api.sendgrid.com/v3/mail/send"
    imapHost: "imap.gmail.com"
    imapPort: 993
    imapUser: "you@gmail.com"
    fromAddress: "bot@example.com"
    ownerEmail: "you@gmail.com"
    classification: CONFIDENTIAL # Default: CONFIDENTIAL

# ---------------------------------------------------------------------------
# Classification: Data sensitivity model
# ---------------------------------------------------------------------------
classification:
  mode: personal # "personal" or "enterprise" (coming soon)
# Levels: RESTRICTED > CONFIDENTIAL > INTERNAL > PUBLIC

# ---------------------------------------------------------------------------
# Policy: Custom enforcement rules (enterprise escape hatch)
# ---------------------------------------------------------------------------
policy:
  rules:
    - id: block-external-pii
      hook: PRE_OUTPUT
      priority: 100
      conditions:
        - type: recipient_is
          value: EXTERNAL
        - type: content_matches
          pattern: "\\b\\d{3}-\\d{2}-\\d{4}\\b" # SSN pattern
      action: REDACT
      message: "PII redacted for external recipient"

    - id: rate-limit-browser
      hook: PRE_TOOL_CALL
      priority: 50
      conditions:
        - type: tool_name
          value: browser
        - type: rate_exceeds
          value: 10/minute
      action: BLOCK
      message: "Browser tool rate limit exceeded"

# ---------------------------------------------------------------------------
# MCP Servers: External tool servers
# ---------------------------------------------------------------------------
mcp_servers:
  filesystem:
    command: "deno"
    args: ["run", "--allow-read", "--allow-write", "mcp-filesystem-server.ts"]
    classification: INTERNAL

  github:
    command: "npx"
    args: ["-y", "@modelcontextprotocol/server-github"]
    classification: CONFIDENTIAL

# ---------------------------------------------------------------------------
# Plugins: Dynamic plugin configuration (optional)
# ---------------------------------------------------------------------------
# ~/.triggerfish/plugins/ मधील Plugins येथे enabled असल्यावर startup वर loaded होतात.
# Agent द्वारे runtime वर loaded plugins (plugin_install द्वारे) ला
# config entry आवश्यक नाही -- ते sandboxed trust आणि manifest classification ला default होतात.
plugins:
  weather:
    enabled: true
    classification: PUBLIC
    trust: sandboxed # किंवा full Deno permissions grant करण्यासाठी "trusted"
    # Additional keys plugin ला context.config म्हणून passed होतात
    api_key: "secret:plugin:weather:apiKey"

  system-info:
    enabled: true
    classification: PUBLIC
    trust: trusted # manifest आणि config दोन्ही "trusted" म्हणायला हवेत

# ---------------------------------------------------------------------------
# Scheduler: Cron jobs and triggers
# ---------------------------------------------------------------------------
scheduler:
  cron:
    jobs:
      - id: morning-briefing
        schedule: "0 7 * * *" # 7 AM daily
        task: "Prepare morning briefing with calendar, unread emails, and weather"
        channel: telegram
        classification: INTERNAL

      - id: pipeline-check
        schedule: "0 */4 * * *" # Every 4 hours
        task: "Check Salesforce pipeline for changes and notify if significant"
        channel: slack
        classification: CONFIDENTIAL

      - id: pr-review-check
        schedule: "*/15 * * * *" # Every 15 minutes
        task: "Check open PR tracking files and query GitHub for new reviews"
        classification: INTERNAL

  trigger:
    interval: 30m # दर 30 minutes check करा
    classification: INTERNAL # Triggers साठी Max taint ceiling
    quiet_hours: "22:00-07:00" # या hours दरम्यान suppress करा

# ---------------------------------------------------------------------------
# Notifications: Delivery preferences
# ---------------------------------------------------------------------------
notifications:
  preferred_channel: telegram # Default delivery channel
  quiet_hours: "22:00-07:00" # Normal/low priority suppress करा
  batch_interval: 15m # Low-priority notifications batch करा

# ---------------------------------------------------------------------------
# Agents: Multi-agent routing (optional)
# ---------------------------------------------------------------------------
agents:
  default: personal # Fallback agent
  list:
    - id: personal
      name: "Personal Assistant"
      channels: [whatsapp, telegram]
      tools:
        profile: "full"
      model: claude-opus-4-5
      classification_ceiling: INTERNAL

    - id: work
      name: "Work Assistant"
      channels: [slack, email]
      tools:
        profile: "coding"
        allow: [browser, github]
      model: claude-sonnet-4-5
      classification_ceiling: CONFIDENTIAL

# ---------------------------------------------------------------------------
# Voice: Speech configuration (optional)
# ---------------------------------------------------------------------------
voice:
  stt:
    provider: whisper # whisper | deepgram | openai
    model: base # Whisper model size
  tts:
    provider: elevenlabs # elevenlabs | openai | system
    voice_id: "your-voice-id"
  wake_word: "triggerfish"
  push_to_talk:
    shortcut: "Ctrl+Space"

# ---------------------------------------------------------------------------
# Webhooks: Inbound event endpoints (optional)
# ---------------------------------------------------------------------------
webhooks:
  endpoints:
    - id: github
      path: /webhook/github
      # Webhook secret OS keychain मध्ये stored आहे
      classification: INTERNAL
      actions:
        - event: "pull_request.opened"
          task: "Review PR and post summary"
        - event: "pull_request_review"
          task: "A PR review was submitted. Read tracking file, address feedback, commit, push."
        - event: "pull_request_review_comment"
          task: "An inline review comment was posted. Read tracking file, address comment."
        - event: "issue_comment"
          task: "A comment was posted on a PR. If tracked, address feedback."
        - event: "pull_request.closed"
          task: "PR closed or merged. Clean up branches and archive tracking file."
        - event: "issues.opened"
          task: "Triage new issue"

# ---------------------------------------------------------------------------
# GitHub: GitHub integration settings (optional)
# ---------------------------------------------------------------------------
github:
  auto_merge: false # Default: false. Approved PRs auto-merge करण्यासाठी true set करा.

# ---------------------------------------------------------------------------
# Groups: Group chat behavior (optional)
# ---------------------------------------------------------------------------
groups:
  default_behavior: "mentioned-only"
  overrides:
    - channel: slack
      channel_name: "#ai-assistant"
      behavior: "always"

# ---------------------------------------------------------------------------
# Web: Search and fetch configuration
# ---------------------------------------------------------------------------
web:
  search:
    provider: brave # Search backend (brave is the default)
# API key OS keychain मध्ये stored आहे

# ---------------------------------------------------------------------------
# Remote: Remote access (optional)
# ---------------------------------------------------------------------------
remote:
  tailscale:
    serve: true
    funnel:
      enabled: true
      paths: ["/webhook/*"]
  auth:
# Auth token OS keychain मध्ये stored आहे

Section Reference

models

Key	Type	वर्णन
primary	object	provider आणि model fields सह Primary model reference
primary.provider	string	Provider name (anthropic, openai, google, ollama, lmstudio, openrouter, zenmux, zai)
primary.model	string	Agent completions साठी वापरलेला Model identifier
vision	string	Automatic image description साठी Optional vision model (Image and Vision पहा)
streaming	boolean	Streaming responses enable करा (default: true)
providers	object	Provider-specific configuration (खाली पहा)
failover	string[]	Fallback models ची Ordered list
failover_config.max_retries	number	Failover पूर्वी per provider retries
failover_config.retry_delay_ms	number	Milliseconds मध्ये retries दरम्यान delay
failover_config.conditions	string[]	Failover trigger करणाऱ्या Conditions

channels

प्रत्येक channel key channel type आहे. सर्व channel types default classification level override करण्यासाठी classification field support करतात.

सर्व secrets (tokens, API keys, passwords) OS keychain मध्ये stored आहेत,

या file मध्ये नाही. Credentials securely enter करण्यासाठी triggerfish config add-channel <name> run करा. :::

classification

Key	Type	वर्णन
mode	"personal" किंवा "enterprise"	Deployment mode (coming soon -- currently दोन्ही same classification levels वापरतात)

policy

Hook execution दरम्यान evaluated custom rules. प्रत्येक rule एक hook type, priority, conditions, आणि action specify करतो. Higher priority numbers पहिले evaluated होतात.

mcp_servers

External MCP tool servers. प्रत्येक server त्याला launch करण्यासाठी command, optional environment variables, classification level, आणि per-tool permissions specify करतो.

plugins

Dynamic plugin configuration. प्रत्येक key ~/.triggerfish/plugins/ मधील directory शी matching plugin name आहे. Configuration optional आहे -- agent द्वारे runtime वर (via plugin_install) loaded plugins config entry शिवाय काम करतात.

Key	Type	Default	वर्णन
enabled	boolean	false	Startup वर हा plugin load करायचा का
classification	string	from manifest	Plugin चे classification level override करा
trust	"sandboxed" किंवा "trusted"	"sandboxed"	Trust level grant. Manifest आणि config दोन्ही "trusted" म्हणायला हवेत
(other keys)	any	--	Plugin ला context.config म्हणून passed होतात

Details साठी Plugins पहा.

scheduler

Cron job definitions आणि trigger timing. Details साठी Cron and Triggers पहा.

notifications

Notification delivery preferences. Details साठी Notifications पहा.

web

Key	Type	वर्णन
web.search.provider	string	web_search tool साठी Search backend (currently: brave)

Details साठी Web Search and Fetch पहा.

logging

Key	Type	Default	वर्णन
level	string	"normal"	Log verbosity: quiet (errors only), normal (info), verbose (debug), debug (trace)

Log output आणि file rotation साठी Structured Logging पहा.

github

Key	Type	Default	वर्णन
auto_merge	boolean	false	true असल्यावर, approving review receive केल्यावर agent PRs auto-merge करतो. false (default) असल्यावर, agent owner ला notify करतो आणि explicit merge instruction साठी wait करतो.

Full setup instructions साठी GitHub Integration guide पहा.