Adding voice support.#55
Merged
Merged
Conversation
All CompressionConfig fields are now configurable via environment variables: - BPSA_COMPRESSION_ENABLED (bool, default: 1) - BPSA_COMPRESSION_KEEP_RECENT_STEPS (int, default: 40) - BPSA_COMPRESSION_MAX_UNCOMPRESSED_STEPS (int, default: 50) - BPSA_COMPRESSION_KEEP_COMPRESSED_STEPS (int, default: 80) - BPSA_COMPRESSION_MAX_COMPRESSED_STEPS (int, default: 120) - BPSA_COMPRESSION_TOKEN_THRESHOLD (int, default: 0 = disabled) - BPSA_COMPRESSION_MODEL (str, model id, default: same as main model) - BPSA_COMPRESSION_MAX_SUMMARY_TOKENS (int, default: 50000) - BPSA_COMPRESSION_PRESERVE_ERROR_STEPS (bool, default: 0) - BPSA_COMPRESSION_PRESERVE_FINAL_ANSWER_STEPS (bool, default: 1) - BPSA_COMPRESSION_MIN_CHARS (int, default: 4096) Changes: - bp_utils.py: add get_env_bool_default() helper - bp_thinkers.py: DEFAULT_THINKER_COMPRESSION reads from env vars - bp_cli.py: build_model() supports override_model_id; build_agent() resolves BPSA_COMPRESSION_MODEL; updated env var documentation - bp_ad_infinitum.py: updated env var documentation Generated by claude-sonnet-4.6 Co-Authored-By: bpsa2 <241537330+bpsa2@users.noreply.github.com>
get_env_bool(name, default=False) now handles both cases: - When called without default, behaves as before (returns False if unset) - When called with default, returns that default if unset Removed redundant get_env_bool_default(). Updated all call sites in bp_thinkers.py to use get_env_bool() with explicit defaults. Generated by claude-sonnet-4.6 Co-Authored-By: bpsa2 <241537330+bpsa2@users.noreply.github.com>
- CLI.md: added 'Context Compression Variables' section with full table of all 11 BPSA_COMPRESSION_* vars, their types, defaults and descriptions - README.md: added a note in the bpsa env vars section pointing to CLI.md for compression vars; added a note at the end of the Context Compression Configuration Options table pointing to CLI.md for env var configuration Generated by claude-sonnet-4.6 Co-Authored-By: bpsa2 <241537330+bpsa2@users.noreply.github.com>
Updated all occurrences of --gui to --gui-x11 in source code and documentation. The argparse attribute args.gui was also updated to args.gui_x11 accordingly. Co-Authored-By: bpsa2 <241537330+bpsa2@users.noreply.github.com>
When no instruction files (e.g. CLAUDE.md, AGENTS.md) are found in the current working directory, load_agent_instructions() now falls back to loading a README file for general project context. Supported formats (in priority order): .md, .txt, .rst, and no extension. Matching is case-insensitive. Content is truncated at 20,000 characters and wrapped in <readme> tags to distinguish it from instruction <filenotes>. Also add tests/test_bp_load_instructions.py with full coverage of the new fallback behaviour (priority, truncation, empty file, case insensitivity, extension preference, etc.). Co-Authored-By: bpsa2 <241537330+bpsa2@users.noreply.github.com>
- pyproject.toml: add optional `voice` extra with voicelistener dependency; include voice in the `all` extras group - bp_cli.py: implement /voice [on|off] slash command backed by VoiceListener; support WhisperTranscriber via BPSA_VOICE_TRANSCRIBER env var; drain voice queue into prompt_toolkit buffer before each render; document BPSA_VOICE_TRANSCRIBER and BPSA_VOICE_MODEL env vars; clean up voice listener on /exit and EOF Co-Authored-By: bpsa2 <241537330+bpsa2@users.noreply.github.com>
Add elevenlabs as a supported voice transcriber alongside whisper, and document the voice input feature in README.md and CLI.md. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The voice listener was being stopped/started on every agent run, creating a new thread each time. This caused threads to accumulate across turns, especially after each final_answer. Fix: remove the stop/start around agent.run(). The voice listener now runs continuously; stale audio captured during agent execution is already discarded by the existing queue drain in get_input(). Model: claude-sonnet-4.6 Co-Authored-By: bpsa2 <241537330+bpsa2@users.noreply.github.com>
Updated by claude-sonnet-4.6 Co-Authored-By: bpsa2 <241537330+bpsa2@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.