## Problem
The CLI package preserved a generic adapter layer and managed Codex
parser path that did not match the supported customer workflows.
Messaging runs Claude Code sessions, while Codex is supported through
`fcc-codex` and extensions.
## Changes
| Before | After |
| --- | --- |
| `fcc-claude` and `fcc-codex` shared generic adapter plumbing. |
`fcc-claude` and `fcc-codex` use explicit launcher modules. |
| Messaging depended on a generic CLI session abstraction. | Messaging
depends on managed Claude Code sessions. |
| Codex catalog generation lived as a top-level CLI helper. | Codex
catalog generation lives under the Codex launcher owner. |
| Tests asserted deleted internal adapter shapes. | Tests assert
launcher, managed-Claude, and customer-surface behavior. |
Pass parent_session_id into get_or_create_session so reply nodes align with
the fork/resume path instead of always allocating a fresh pending session.
Add unit coverage and update integration expectations.
Consolidates the incremental refactor work into a single change set: modular web tools (api/web_tools), native Anthropic request building and SSE block policy, OpenAI conversion and error handling, provider transports and rate limiting, messaging handler and tree queue, safe logging, smoke tests, and broad test coverage.
…x cancel_all() TOCTOU
- Remove tree_queue property setter (backward-compat hack; all callers
already migrated to replace_tree_queue()); keep property getter only
- Update 2 remaining tests that still used direct assignment to use
replace_tree_queue()
- Remove _set_connected() 1-line wrapper on DiscordPlatform; assign
_connected directly
- Fix cancel_all() TOCTOU: hold self._lock for the full loop so newly
created trees cannot slip through between the snapshot and cancellation
---------
Co-authored-by: Claude <noreply@anthropic.com>
## Summary
Added NVIDIA NIM as a second transcription option ( alongside local
Whisper). This lets you transcribe voice notes using NVIDIA's cloud API
instead of running Whisper locally.
## What changed
- **Transcription**: Now supports the two backends
- Local Whisper: Free, runs on your GPU/CPU (existing)
- NVIDIA NIM: Cloud API via Riva gRPC (new)
- **Supported models**: 8 NVIDIA NIM models added (Parakeet variants for
different languages, Whisper Large V3)
---------
Co-authored-by: Alishahryar1 <alishahryar2@gmail.com>
The max_sessions cap in CLISessionManager was the only thing enforcing
a limit on concurrent CLI processes. Now that provider concurrency is
controlled at the streaming layer (PROVIDER_MAX_CONCURRENCY semaphore),
the CLI session pool cap is redundant and removed entirely.
Changes:
- cli/manager.py: remove max_sessions param, cap check, _cleanup_idle_sessions_unlocked, max_sessions from get_stats()
- config/settings.py: remove max_cli_sessions field
- api/app.py: remove max_sessions=settings.max_cli_sessions from CLISessionManager constructor
- messaging/handler.py: remove "Waiting for slot" status check; stats display no longer shows Max CLI
- .env.example: remove MAX_CLI_SESSIONS line
- tests/cli/test_cli.py: remove max_sessions args and assertion from manager tests
- tests/cli/test_cli_manager_edge_cases.py: remove two tests for cap/cleanup behavior
- tests/api/test_app_lifespan_and_errors.py: remove max_cli_sessions from all SimpleNamespace settings
- tests/config/test_config.py: remove max_cli_sessions isinstance assertion
- tests/conftest.py: remove max_sessions from mock stats
- tests/messaging/test_handler.py: merge slot/capacity tests into single new-conversation test; remove Max CLI assertion from stats test
- tests/messaging/test_handler_markdown_and_status_edges.py: remove "Waiting for slot" assertion; drop max_sessions from all stats mocks
https://claude.ai/code/session_014mrF1WMNgmNjtPBuoQHsbg
- Introduced message_thread_id to the IncomingMessage model for handling forum topic IDs in Telegram.
- Updated messaging platforms (Discord and Telegram) to accept and process message_thread_id in send_message methods.
- Modified message handlers to utilize message_thread_id when sending messages.
- Enhanced test cases to validate the integration of message_thread_id in message handling.
This change improves support for forum supergroups in Telegram and enhances message management across platforms.
- Implemented functionality to cancel pending voice transcriptions when a user replies with the /clear command.
- Updated the Telegram and Discord platform classes to manage pending voice messages, including registration and cancellation logic.
- Enhanced the message handler to delete associated messages and notify users when a voice note is cancelled.
- Added tests to ensure the cancellation feature works as expected during transcription.
- Updated transcription logic to utilize Hugging Face's Whisper models instead of faster-whisper.
- Introduced new model mapping and pipeline loading functions.
- Adjusted tests to reflect changes in the transcription process.
- Updated documentation in README, .env.example, and settings to align with the new implementation.
- Ensured compatibility with CUDA 13 and removed unnecessary dependencies.
- Validate whisper_device in Settings and _get_local_model
- Reject 'auto' with clear ValueError/ValidationError
- Update docs in config, .env.example, README
- Add tests for invalid device and valid cpu/cuda
Co-authored-by: Ali Khokhar <alishahryar2@gmail.com>
- Remove _cuda_failed_models and inference-time CPU fallback
- auto: try CUDA only, fail fast on RuntimeError (no CPU fallback)
- cpu/cuda: use device directly, fail fast on errors
- Update docs in config, .env.example, README
Co-authored-by: Ali Khokhar <alishahryar2@gmail.com>
- Introduced a new optional field `status_message_id` in the IncomingMessage model to track the status of voice note processing.
- Updated the Telegram and Discord platforms to utilize the `status_message_id` for editing status messages instead of sending new ones.
- Modified tests to assert the correct status message ID is used during voice note handling.
- Changed status message text from "Processing voice note..." to "Transcribing voice note..." for clarity.
- Implemented a queue message to indicate "Processing voice note..." for both Telegram and Discord platforms.
- Updated the Telegram platform to send a status message when handling voice notes.
- Enhanced the test for Telegram voice handling to verify the queue message is sent correctly.
- Introduced a mechanism to track models that fail CUDA inference, allowing for automatic fallback to CPU on subsequent requests.
- Updated the model loading logic to respect the CUDA failure state, improving robustness in audio transcription.
- Removed debug logging code from Discord platform to streamline error handling.
- Introduced voice note handling for Discord and Telegram platforms.
- Added configuration options for voice note functionality in settings.py and .env.example.
- Updated README to include voice note instructions and configuration details.
- Implemented audio attachment processing and transcription using faster-whisper.
- Enabled voice note support through message handlers in both platforms.
- Add MessageTree.set_current_task() method
- Update tree_processor to use set_current_task instead of _current_task
- Move nim_settings out of ProviderConfig, pass only to NvidiaNimProvider
- Update api/dependencies and all tests
Co-authored-by: Ali Khokhar <alishahryar2@gmail.com>
- Implemented handling of the `/clear` command to clear specific branches or entire trees based on message replies.
- Added tests for various scenarios of the clear command, including clearing branches, handling unknown replies, and clearing entire trees.
- Enhanced `TreeQueueManager` with methods to cancel branches and remove subtrees, ensuring proper state management in the session store.
- Updated `SessionStore` and `TreeRepository` to support removal of node mappings and trees, improving data integrity during clear operations.
Ali Khokhar