Transcription Analysis

What it does

When your agent uses it

• Tasks involving transcription providers
• Caption streaming
• The audio pipeline

Example prompts

• “/transcription-analysis”

Requirements

• Python 3

Workflow steps

10 steps, taken from the first numbered list in SKILL.md.

1. 1Create portal/transcription/providers/{name}.py
2. 2Implement class inheriting TranscriptionProvider from providers/base.py
3. 3Implement process_chunk(chunk, language_code, model_variant, config, booth_state) -> str
4. 4For streaming (WebSocket-based): override run_stream(process, language_code, model_variant, config, broadcast_callback, booth_id) — see…
5. 5Add ProviderEnum.{NAME} = "{name}" in constants.py
6. 6Add {name}: {allowed_models_set} to ALLOWED_MODELS
7. 7Add to PROVIDERS dict in worker.py
8. 8If API key needed: add encrypted column to Event model (migration 008 pattern) + update get_api_key key_map in base.py
9. 9Update admin_event_api_settings_post in portal/routers/admin/settings.py to handle new key
10. 10Update templates/admin/api_settings.html

What it can do on your machine

Read from SKILL.md and the folder at commit 1ca0139. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • uv

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use uv, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Transcription Analysis loads about 1.5k tokens when it runs. Until then it costs about 55 tokens; SKILL.md has 455 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~55

When it runs · the whole SKILL.md, loaded when a task matches

~1.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from fossasia/eventyay-interpretation at commit 1ca0139, republished under its Apache-2.0 licence (© fossasia). 455 words, ~1,475 tokens.

name

transcription-analysis

description

Use this skill for tasks involving transcription providers, caption streaming, or the audio pipeline. Reference: `portal/transcription/`, [TRANSCRIPTION_MAP.md](../../context/TRANSCRIPTION_MAP.md).

Interpreter browser
  └── getUserMedia (mic, 16-bit PCM via WebRTC/WHIP)
        └── MediaMTX (RTSP port 8554, path: {event_slug}/{language_code})
              └── ffmpeg (spawned by worker.py)
                    -rtsp_transport tcp -i rtsp://mediamtx:8554/{path}
                    -f s16le -acodec pcm_s16le -ar 16000 -ac 1 -
                        └── TranscriptionProvider.run_stream()
                              └── 3-second PCM chunks
                                    └── process_chunk() → text
                                          └── CaptionAggregator
                                                └── broadcast_transcription()
                                                      ├── /ws/booth/{booth_id}  (interpreters)
                                                      └── /ws/captions/{booth_id}  (listeners)

aggregator = CaptionAggregator(broadcast_callback)
await aggregator.handle_partial(booth_id, partial_text)   # streaming partial
await aggregator.handle_final(booth_id, final_text)       # completed utterance
await aggregator.handle_chunk(booth_id, whisper_chunk)    # Whisper-style chunks
await aggregator.handle_clear(booth_id)                   # silence endpoint

© fossasia, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .agents/skills/transcription-analysis of fossasia/eventyay-interpretation.

Open the folder on GitHubat commit 1ca0139

Used in 1 other repository

We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders. This page covers the copy in fossasia/eventyay-interpretation, which our catalogue first saw on October 7, 2026.

• fossasia/voxbento.agents/skills/transcription-analysis