Track PIDs of backends started by /backend command and only kill those
specific PIDs. Previous implementation used pkill which would kill all
Ollama/llama-server processes including those started by the user or
systemd services. Now we only terminate processes we started.
- Add local_providers with ollama and llamacpp configurations
- /backend command now stops current daemon before starting new one
- Start backends as detached processes to avoid blocking TUI
- Wait 500ms for daemon to initialize before switching
- Add session_config SQLite table for per-session settings
- Update routing to support session override → agent config → global default resolution chain
- Upgrade WebChat SessionBridge from NativeAgent to AgentOrchestrator
- Add /model, /local, /cloud commands to Telegram adapter
- Add /model command to WebChat gateway handlers
- Clear session overrides on /reset command
- Pass memoryStore and config through to SessionBridge
- Add comprehensive tests for all new functionality
Fixes model persistence bug where TUI model changes didn't affect WebChat/Telegram sessions. Now:
- TUI /model sets global default (persists across restarts, affects all new sessions)
- WebChat/Telegram /model sets session override (only that conversation, cleared on /reset)
- WebChat sessions gain AgentOrchestrator features (delegation, compaction, memory)
- README.md: Update audio config format to match schema (enabled + provider.* fields instead of old transcription_endpoint fields), add whisper.cpp server Docker example
- CHANGELOG.md: Add '### Fixed' section with voice message failure handling details
- config/default.yaml: Update audio section with new schema format and Docker setup example
- Send user feedback when voice/audio download fails instead of silent failure
- Send graceful message when audio transcription is not configured instead of empty text which crashes API
- Add capabilities.test.ts (18 tests) for supportsAudioInput()
- Add 15 audio tests to media.test.ts (hasAudio, stripAudioParts, attachmentToAudioSource)
- Add estimateAudioTokens() to tokens.ts (base64→bytes→duration→tokens)
- Update estimateMessageTokens() to include audio content parts
- Add 5 audio token tests to tokens.test.ts
- Add supports_audio config override to model schema
- Wire supports_audio from tier config through routing to capability check
Total tests: 1369 (was 1331, +38 audio-related)
- Create capabilities.ts with supportsAudioInput() detection
- Gemini, OpenAI, and GitHub Models get native audio passthrough
- Anthropic, Bedrock, Ollama, llama.cpp fall back to Whisper transcription
- routing.ts now checks model capability before deciding to transcribe
- Audio attachments are stripped for non-native models (only transcript text passed)
- Remove deprecated audioConfig from createMessageRouter deps (read from config.audio)
- Add AudioSource interface and 'audio' variant to MessageContentPart union
- Update buildUserMessage() to create audio content parts from attachments
- Add attachmentToAudioSource(), hasAudio(), stripAudioParts() helpers
- Gemini: native audio via inlineData (same format as images)
- OpenAI/GitHub: native audio via input_audio content parts
- Anthropic/Bedrock: graceful fallback to transcript text
- Update getMessageTextWithTools() to handle audio blocks for local models
- Add createAudioTranscribeTool with OpenAI/Groq/Ollama/llama.cpp provider support
- Refactor audio config schema to nested audio.enabled + audio.provider structure
- Move audio tool registration to initTools() for conditional enablement
- Fix duplication bug in audio-transcribe.ts URL download handler
- Support base64 data and URL-based audio input with format detection
- Add curly braces to all if/else/for/while statements
- Fix indentation and trailing spaces
- Auto-fixed 372 linting errors using eslint --fix
- Remaining issues are warnings only (non-null assertions, explicit any types)
- Add eslint.config.js using new flat config format
- Configure @typescript-eslint/parser and plugin for TypeScript files
- Add separate config for vanilla JavaScript files (gateway/ui)
- Include Node.js and browser globals
- Enable strict rules: curly braces, no-eval, eqeqeq, etc.
- Configure TypeScript-specific rules (no-explicit-any, no-non-null-assertion)
- Add @typescript-eslint/parser and @typescript-eslint/eslint-plugin dependencies
- Use pnpm for all build, dev, test, and quality check commands
- Replace manual PID file handling with systemd service control
- Add daemon-start, daemon-stop, daemon-restart, daemon-status, daemon-logs targets
- Add enable/disable targets for boot startup management
- Provide convenience aliases (stop, restart, status, logs) for common operations
- Integrate with existing flynn.service systemd user service
tsc only compiles .ts files — the webchat static files (HTML, CSS, JS)
in src/gateway/ui/ were never copied to dist/gateway/ui/, causing 404s
when running the production build via 'pnpm start'.
OpenAI-compatible providers return 'stop' and 'tool_calls' as finish_reason
values, but Flynn's agent loop expects Anthropic-style 'end_turn' and
'tool_use'. This caused the agent to exit the tool loop prematurely when
falling back to GitHub Copilot (due to Anthropic API quota exhaustion).
- openai.ts: Map 'stop' → 'end_turn', 'length' → 'max_tokens', tool_calls
with actual tools → 'tool_use', tool_calls without tools → 'end_turn'
- github.ts: Handle edge case where finish_reason is 'tool_calls' but no
tools were parsed
- agent.ts: Accept both 'tool_use' and 'tool_calls' as valid stop reasons
(belt-and-suspenders), extract toolCalls to local variable for TS narrowing
- openai.test.ts: Update expectations to match new normalized values
- agent.ts: track consecutive calls to the same tool (ignoring args) and
inject a nudge after 4 repeats telling the model to summarize and respond,
preventing local models from endlessly retrying searches with slight
query variations
- agent.ts: wrap the entire tool loop iteration in try-catch so model/network
errors don't crash the daemon — returns a descriptive error message instead
- Tests for both: nudge triggers after 4 same-tool calls, error recovery
persists to history
- ollama.ts: add normalizeMessagesForOllama() converting Anthropic-style
tool_use/tool_result blocks to Ollama's native tool_calls + role:tool format
- llamacpp.ts: add normalizeMessagesForLlamaCpp() with hybrid approach —
assistant tool_calls in native format, but tool results as structured user
messages (many GGUF templates silently drop role:tool messages)
- llamacpp.ts: add configurable requestTimeout with AbortController (default 3min)
- Both use fast-path when no tool blocks are present (zero overhead)
- Full test coverage for both normalizers: plain text passthrough, tool_use
conversion, tool_result mapping, multi-tool round trips, error results
- SOUL.md: list all available tools (web.search, memory.*, cron.*, etc.)
and add Tool Usage Rules section enforcing 'act, don't narrate'
- cron.ts: add getJob(), addJob(), removeJob() to CronScheduler for
runtime (ephemeral) cron job management
- cron tools: add cron.create and cron.delete tools, enhance cron.list
to show schedule/output/message details
- policy.ts: add cron tools to messaging and coding profiles, add
group:cron to tool groups
Fixes issue where models would narrate tool intent ('let me search...')
then stop without actually calling tools.
Allow cron jobs to specify a `model_tier` field that controls which LLM
tier handles the job, without needing separate agent configs. Precedence:
cron job model_tier > agent config > global primary_tier > 'default'.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Local backends using strict chat templates (e.g. Mistral 3) rejected
Flynn's Anthropic-style tool_use/tool_result content blocks, causing
'roles must alternate' errors. Added getMessageTextWithTools() and
normalizeMessagesForLocal() to serialize structured blocks to plain
text, drop empty messages, and merge consecutive same-role messages.
Also fixed compaction to ensure kept messages start with user role.
Adds process.loadEnvFile() to CLI entry point so API keys (ZHIPUAI_API_KEY,
OPENROUTER_API_KEY, XAI_API_KEY, etc.) can be stored in a project .env file
instead of shell environment or systemd service config. Uses Node >= 20.12
built-in — no dotenv dependency needed. Silent no-op if .env doesn't exist.
Updates .env.example with placeholders for all provider API keys.
Previously, switching to zhipuai/openrouter/xai via /model would throw a
confusing 'OPENAI_API_KEY missing' error from the OpenAI SDK. Now
createClientFromConfig validates API keys before constructing the client,
throwing errors that name the correct env var (e.g. ZHIPUAI_API_KEY).
Also fixes the misleading 'as anthropic' type cast in the /model handler
to validate against MODEL_PROVIDERS and use the ModelProvider type.
Wrap each message in a .message-wrapper div and render copy/edit buttons
below the bubble instead of overlapping inside it. Improves readability
and prevents buttons from covering message content.
Extract MODEL_PROVIDERS const from config schema as single source of truth
for provider names. PROVIDER_NAMES in TUI commands now imports from schema
instead of maintaining a hardcoded list. Adds tests verifying sync.
Updates README TUI Commands section with /model hot-swap documentation,
supported providers, and runtime model switching examples.
Copy button on all messages (clipboard API with checkmark feedback).
Edit button on user messages populates the input textarea.
Buttons appear on hover (desktop) or always visible (mobile).
ModelRouter now supports multiple tier-change listeners via addOnTierChange(),
SessionBridge subscribes to tier changes and propagates them to all WebChat
agents (both existing and newly created), and the fullscreen TUI now also
updates the agent's tier when switching models (matching minimal TUI behavior).
Local LLMs often get stuck calling the same tool repeatedly because they
lack the sophistication to synthesize results. The agent loop had no
safeguard — it re-executed whatever the model requested up to 10 times.
Add fingerprint-based loop detection: if the same tool+args combination
repeats 3 consecutive times, break the loop and return the last results.
Also add agents.max_iterations to the config schema so the iteration
limit is user-configurable (default: 10).
Gmail API returns snippets with HTML entities (&, ', <br>, etc.)
that leaked into LLM responses as raw HTML. Added shared sanitizeHtml()
utility in src/utils/html.ts and applied it to gmail tool snippets,
HTML body fallback, and gmail watcher snippets.
Use agent.reset() instead of session.clear() when agent is available,
ensuring tool fingerprint, usage counters, and agent history are all
cleared properly alongside the session.
Replace process.stdout.write-based onToolUse callback (which corrupts
Ink rendering) with a React state-driven approach that shows tool names,
args, and completion status in the streaming content area.
Fullscreen TUI was calling modelClient directly, bypassing the NativeAgent
tool loop entirely. Pass the agent through FullscreenTuiConfig → App and
use agent.process() for message handling, which enables the full tool
registry and executor.
Add three new Google service integrations following the established
Gmail/GCal pattern:
- Google Docs (docs.list, docs.search, docs.read): list, search, and
read document content as plain text via Docs + Drive APIs
- Google Drive (drive.list, drive.search, drive.read): list, search,
and read files with export support for Workspace files (Docs→text,
Sheets→CSV, Slides→text)
- Google Tasks (tasks.lists, tasks.list): list task lists and tasks
with status, due dates, and notes
Each service has its own config section, OAuth auth command, tool
policy group, and test suite (53 new tests). The setup wizard now
offers to configure all Google services together and run OAuth auth
flows automatically after saving config.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Persist /model tier choice to ~/.local/share/flynn/preferences.json so
it survives restarts. Decode HTML entities (e.g. ') in markdown
renderer output. Suppress noisy logger.info and punycode deprecation
warnings in TUI startup.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The existing gmail.list and gmail.search tools only return snippets.
gmail.read fetches the full message by ID using format: 'full', decodes
base64url body parts (preferring text/plain, falling back to stripped
HTML), and returns headers + body text. This enables workflows like
searching for invoices and extracting amounts from the full content.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add GCal tools setup guide to README (prerequisites, config, fields).
Add gmail-auth, gcal-auth, setup to the CLI commands table. Add
"Adding a New Tool" checklist to CLAUDE.md covering the full wiring
chain including the TUI registration gotcha.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
William Valentin