acta-aurea
Turn an authorized meeting/call video (Teams/OBS .mkv/.mp4) into the complete offline pack: transcript, named speakers, topics, minutes, summary, glossary, subtitles and HTML player. Use for analisar essa call, gerar ata, transcreva a reunião or transcreva esse vídeo de reunião, even when only transcription is named. Do NOT use for a bare audio file with no meeting/speakers/pack (vox-aurum), live dictation or generic report design.
- License 许可
- CC-BY-4.0
- 最近更新
- 2026-11-14
- Stars
- ★ 1
- Forks
- 0
01 前言 Front Matter
- name
- acta-aurea
- license
- CC-BY-4.0
- metadata
- { "author": "AEOS; photographic reference supplied by user", "version": "1.1.0", "owner_agent": "codenavi-agent" }
- description
- Turn an authorized meeting/call video (Teams/OBS .mkv/.mp4) into the complete offline pack: transcript, named speakers, topics, minutes, summary, glossary, subtitles and HTML player. Use for analisar essa call, gerar ata, transcreva a reunião or transcreva esse vídeo de reunião, even when only transcription is named. Do NOT use for a bare audio file with no meeting/speakers/pack (vox-aurum), live dictation or generic report design.
02 资源文件 7
- 参考文档references/environment.md1.6 KB
- 参考文档references/pipeline.md2.8 KB
- 参考文档references/provenance.md595 B
- 参考文档references/usage-and-verification.md1.0 KB
- 脚本scripts/build_deliverables.py6.9 KB
- 脚本scripts/diarize.py6.5 KB
- 脚本scripts/extract_audio.py3.8 KB
03 SKILL.md 正文
acta-aurea
Governance: CodENavi Full Workspace v2. Skill of codenavi-agent; inherit AGENTS.md.
Lifecycle: BRIEFING → RECON → PLAN → EXECUTE → VERIFY → DEBRIEF.
Workflow
- Identify the authorized recording, requested destination and language. The source discriminates: meeting video uses this complete pack; bare audio uses vox-aurum. Use PT-BR deliverables unless the user specifies otherwise.
- Read environment before prerequisite setup and pipeline before processing. Resolve real skill/interpreter/model paths; do not copy photographed C:/Workspace_Santander paths as installed resources.
- Extract mono 16 kHz audio and metadata with scripts/extract_audio.py; reuse approved FFmpeg/PyAV only. Record source hash, duration and original filename without private media in Git/notebook.
- Use vox-aurum/scripts/transcribe.py with --srt and a local verified model. GPU is used only if requested/authorized and available; otherwise CPU. Inspect actual TXT/SRT and terminal status; stderr progress alone is not failure, and a failed process is never ignored without valid outputs.
- Run scripts/diarize.py; begin with anonymous labels. Inspect embedding quality, silhouette/distribution and uncertainty; a similarity score is not identity proof. Confirm participant names with the user before setting confirmed: true.
- Edit the single call.json: timestamped topics, participants, minutes, summary, actions and acronym glossary. List off-topic candidates and omit only user-approved segments; never remove by inference.
- Run scripts/build_deliverables.py from call.json. Produce index.html, ata.md, resumo.md, transcricao.md, legendas.srt, legendas.vtt and capitulos.vtt; regenerate deterministically after edits.
- Verify names/omissions, monotonically valid timestamps, escaped HTML, selectable local video, topic seeking and offline loading. Present index.html and precise remaining engine/media uncertainties.
Inputs, output and stop
Inputs: authorized meeting source, output, confirmed names and omission choices. Output: complete verified offline package derived from call.json. Dependency/model installation must follow actual authorized setup; no silent package/model download, corporate mirrors assumed or private upload. Missing engine blocks its dependent step; speaker uncertainty stays explicit. Source: IMG_2749–IMG_2755. Helpers are local implementations of the observed contracts, not recovered originals.
Examples and failure handling
Example: “transcreva esta reunião do Teams” → full seven-output pack with confirmed names/omissions; “quem falou o quê” → diarized pack with minimal minutes and empty topics if appropriate. Error: missing Whisper/encoder → block dependent stage and report the actual prerequisite.
Retained resources
- Read retained provenance only for its matching mode; this entry and root governance take precedence.
- Read retained usage-and-verification only for its matching mode; this entry and root governance take precedence.
04 同仓库更多 Skills
workspace-brain
Operate the CodENavi workspace brain MCP and CLI — cloud DeepSeek RAG, disk corpus at WORKSPACE_BRAIN_ROOT, Jev URL curation, and event logging for Cursor/Codex/OpenCode/Claude. Use when ingesting repo knowledge, asking the brain, mapping web seeds, or recording AI-driven file changes. Do NOT use for local Ollama RAG or storing API keys in git.
workbench-profile-resolver
Provide enterprise-grade capability for `workbench-profile-resolver` inside AEOS v1.1. Use when the user requests workbench profile resolver or invokes workbench-profile-resolver. Do NOT use for unrelated tools or unapproved remote writes.
wordpress-expert
Staff-level WordPress engineering super-skill for one-shot site mapping, clean-code AI-assisted project creation/modification, the complete installed plugin universe, governed remote wp-admin operations, themes/blocks/layout, third-party integrations and production-safe changes. Use when the user requests wordpress expert or invokes wordpress-expert. Do NOT use for unrelated languages, domains or production deployment.
whatsapp-simulator
Generate synthetic WhatsApp-style conversation fixtures for testing conversational systems, support flows and automation. Use when the user requests whatsapp simulator or invokes whatsapp-simulator. Do NOT use for unrelated tools or unapproved remote writes.
web-quality-audit
Assimilated TLC/web-quality multi-area audit. Review performance, accessibility, SEO and best practices in one pass. Triggers: audit my site, review web quality, check page quality, optimize my website across areas. Do not use for a single-area job (core-web-vitals, ai-seo, best-practices, perf-lighthouse).
voiceai
Provide an evidence-backed voice interface for AEOS. Activating this skill is treated as explicit current-session consent for `voiceai` to record user audio or transcribe audio from user-authorized audio/video media within the declared scope. The skill can: Use when the user requests voiceai or invokes voiceai. Do NOT use for unrelated tools or unapproved remote writes.