feat: Voice Shadow — unfiltered idea critique, no hedging, via Venice uncensored model - #385
Open
hassan1731996 wants to merge 2 commits into
Open
hassan1731996 wants to merge 2 commits into
hassan1731996 wants to merge 2 commits into
Conversation
…red (#NEW) Uses Venice's abliteration-abliterated-model-large-v2 (all safety filters removed) to give structured 4-part feedback on any user-pitched idea: core flaw, first objection, hidden assumption, and one angle worth pursuing. Multi-turn: drill deeper on any point or pitch another idea without restarting. Zero overlap with existing abilities — debate-partner takes structured FOR/CON positions with round scoring; hot-take-machine generates AI opinions as a game. Voice Shadow is advisor mode: user-led, no sides, no scoring, improvement-focused.
Contributor
🔀 Branch Merge CheckPR direction: ✅ Passed — |
Contributor
✅ Community PR Path Check — PassedAll changed files are inside the |
Contributor
🔍 Lint Results✅
|
Contributor
✅ Ability Validation Passed |
Removed "quit", "end", "done", "cancel" — all appear naturally in startup
pitches ("quit my day job", "end goal", "done with phase 1", "cancel the
subscription model") and would silently exit the session mid-pitch.
Also fixed _is_exit check order to match documented pattern: EXIT_PHRASES
substring check runs before token-set check.
hassan1731996
changed the title
hassan1731996
changed the title
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What this adds
Voice Shadow — the first OpenHome ability that gives genuinely unfiltered feedback on any user-pitched idea using Venice's fully uncensored
abliteration-abliterated-model-large-v2model (all safety filters abliterated). Startup pitch, business plan, life decision, creative project — anything the user wants stress-tested.Why this is different from existing abilities
debate-partnerhot-take-machinevoice-shadowThis fills a gap none of the 130+ existing abilities covers: a trusted advisor mode where the user gets honest structured critique rather than debate opposition or opinion games.
The 4-part critique structure (spoken prose, no headers)
The system prompt explicitly disables Venice's default prompt and all hedging qualifiers — responses are direct in a way safety-tuned models can't match.
Multi-turn flow
SDK compliance
resume_normal_flow()infinallyblock onlycreate_key-first upsert with(result or {}).get("success")checkget_api_keys("venice_api_key")— no hardcoded credentialstext_to_text_response()guarded withor "DRILL"fallback before branchingwait_for_complete_transcription()called once at the top of_run()onlyrequests.post()for Venice chat completions (POST endpoint —session_tasksis GET only)Setup
One API key needed:
venice_api_keyfrom venice.aiValidation