[codex] voxcpm phone assistant bridge#345
Draft
jjuniper-dev wants to merge 5 commits into
Draft
Conversation
Author
|
I need Claude to take a look at this and see if architecturally sound and if it's good for a release |
…arity The capture payload emitted by the phone assistant did not match schemas/pca_capture_event.schema.json (additionalProperties: false, required classification/provenance, content not text), so a validating WF10/WF-Dispatch gateway would reject every capture with HTTP 400 and the failure was swallowed silently. Rebuild the payload to the contract: source=iphone_shortcut, capture_type, timestamp, content, classification (env-overridable via PCA_CAPTURE_CLASSIFICATION, default confidential), and provenance. Off-contract keys (text/context_note/metadata) are dropped. Also fix the "Send & Speak" handler: send_btn declared 7 outputs but prepare_turn returned 6, so Gradio raised on every click. Return profile_data as the 7th value. Update tests and README to the conformant contract. Claude-Session: https://claude.ai/code/session_01RwWk3CSfGsNqpdr47PEMFN
Decouple the phone assistant bridge from PCA so it can be upstreamed:
the PCA-specific ingestion schema mapping now lives downstream, not in
this public TTS project.
- Rename env/UI config to neutral names: ASSISTANT_BACKEND_URL/TOKEN,
ASSISTANT_CONTEXT, ASSISTANT_BACKEND_MODE, ASSISTANT_BACKEND_MODEL,
CAPTURE_WEBHOOK_URL/TOKEN. No "PCA" identifiers remain in the code.
- Capture webhook now POSTs a neutral, backend-agnostic event
({source, type, timestamp, message, reply, profile, turn}); mapping it
onto any ingestion schema is the receiving webhook's job.
- OpenAI-compatible backend model name is configurable (default "assistant")
instead of hardcoded "pca".
- Drop the personal "reddit-female" default profile; --default-profile-name
now defaults to empty (no auto-load).
- Mask backend/capture tokens in the UI (type="password").
- Update tests and README to the neutral contract.
Claude-Session: https://claude.ai/code/session_01RwWk3CSfGsNqpdr47PEMFN
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.
Summary
funasr.voxcpm-assistantas a CLI entrypoint.source,capture_type,timestamp, andtextexplicitly.Why
Validation
python -m py_compile src/voxcpm/phone_assistant.py tests/test_phone_assistant.pyuv run voxcpm-assistant --helppytestis not installed in this checkout, so the test file could not be executed viauv run pytesthere.Notes