lai-align SKILL.md skillA

lai-align from lattifai/lattifai-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add lattifai/lattifai-skills/lai-align@git:20260428.4ab9c84

Pins git:20260428.4ab9c84 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
lattifai/lattifai-skills
Path
skills/lai-align/SKILL.md
Versions recorded
5
Latest version
git:20260428.4ab9c84, committed Apr 28, 2026
First version recorded
Apr 12, 2026
Size
3,162 bytes
Repository stars
32
License
MIT

Align existing captions to audio/video with word-level precision using the Lattice-1 model. Trigger when the user has both a media file AND a caption/transcript that need to be synchronized, or says "fix caption timing", "字幕对不上", "对齐字幕", "word-level timestamps", "karaoke timing", "timestamps are off". Do NOT trigger without existing text — use `/lai-transcribe` first.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# LattifAI Forced Alignment

Produces word-level and segment-level timestamps with sub-frame accuracy. Requires an API key — see `/lai-setup`.

## Basic Command

Positional args are `input_media`, `input_caption`, `output_caption`. Pick a `<base>` (the media stem or YouTube ID) and reuse it; outputs land in the current directory:

```bash
# <base> = podcast (from podcast.mp3)
lai alignment align podcast.mp3 podcast.srt podcast.aligned.json
# shortcut:
lai-align podcast.mp3 podcast.srt podcast.aligned.json
# or explicit key=value style (interchangeable, useful inside scripts):
lai alignment align --direct -Y \
    input_media=podcast.mp3 \
    input_caption=podcast.srt \
    output_caption=podcast.aligned.json \
    caption.input.split_sentence=false
```

Format is inferred from the output file extension (30+ formats — see `/lai-caption`). Add `--direct -Y` for non-interactive pipeline runs.

**Output naming**: prefer `<base>.aligned.json` for downstream pipelines (karaoke / translate / diarize all consume aligned JSON). Use `<base>.srt` etc. only when the alignment result is the final deliverable.

## Common Options

Append as `key=value` pairs:
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add lattifai/lattifai-skills/lai-align@sha256:16a2bb9865df88e4

The line at the top of this page pins the label git:20260428.4ab9c84, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_fqsoxr3xgkan3vdz.svg)](https://markdownregistry.com/a/art_fqsoxr3xgkan3vdz)

0 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 5 versions of this file, between Apr 12, 2026 and Apr 28, 2026, about one every 4 days.

versioncommittedcommitsizeaudit
git:20260428.4ab9c84 latest2026-04-28 4ab9c84 3,162 BA view · diff
git:20260417.89d056b2026-04-17 89d056b 8,643 BA view · diff
git:20260416.83f69072026-04-16 83f6907 2,832 BA view · diff
git:20260412.c7d541e2026-04-12 c7d541e 2,183 BA view · diff
git:20260412.257645d2026-04-12 257645d 8,124 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (3162 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

lattifai/lattifai-skills · 32 stars · license MIT · pushed 2026-11-09 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_fqsoxr3xgkan3vdz
GET https://markdownregistry.com/api/v1/resolve?ref=lattifai/lattifai-skills/lai-align
GET https://markdownregistry.com/api/v1/blob/16a2bb9865df88e499b6241adc30b5433e081a1607d580fc648beba6ac062a13

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from lattifai/lattifai-skills

CLAUDE.md CLAUDE.md file
lattifai/lattifai-skills · CLAUDE.md
git:20260428.4ab9c84 · audit A · 32 stars
lai-caption SKILL.md skill
lattifai/lattifai-skills · skills/lai-caption/SKILL.md · Convert between 30+ caption/subtitle formats (SRT, VTT, ASS, JSON, TextGrid, LRC, FCPXML, Premiere, …) and shift…
git:20260428.4ab9c84 · audit A · 32 stars
lai-diarize SKILL.md skill
lattifai/lattifai-skills · skills/lai-diarize/SKILL.md · Identify speakers ("who said what") in aligned captions via pyannote.audio. Real speaker names come from the agent's…
git:20260514.830b73a · audit A · 32 stars
lai-karaoke SKILL.md skill
lattifai/lattifai-skills · skills/lai-karaoke/SKILL.md · Generate viral-ready karaoke / lyric-video subtitles with per-word highlighting in one command. Adaptive font size from…
git:20260428.4ab9c84 · audit A · 32 stars
lai-setup SKILL.md skill
lattifai/lattifai-skills · skills/lai-setup/SKILL.md · Install LattifAI and get a free trial API key. Trigger on first mention of LattifAI / `lai`, authentication errors…
git:20260427.849e193 · audit B · 32 stars
lai-summarize SKILL.md skill
lattifai/lattifai-skills · skills/lai-summarize/SKILL.md · Summarize a transcript, podcast, or long caption file into structured markdown (TL;DR, chapters with timestamps…
git:20260428.4ab9c84 · audit A · 32 stars
lai-transcribe SKILL.md skill
lattifai/lattifai-skills · skills/lai-transcribe/SKILL.md · Transcribe audio/video to timestamped captions with Gemini (100+ languages) or local Parakeet / SenseVoice models…
git:20260428.4ab9c84 · audit A · 32 stars
lai-translate SKILL.md skill
lattifai/lattifai-skills · skills/lai-translate/SKILL.md · Translate captions into another language (or produce bilingual captions) while preserving segment count, timing…
git:20260513.0a51f69 · audit A · 32 stars
lai-youtube SKILL.md skill
lattifai/lattifai-skills · skills/lai-youtube/SKILL.md · One-command YouTube pipeline — downloads media, fetches or transcribes captions, and runs forced alignment. Trigger…
git:20260428.4ab9c84 · audit A · 32 stars

Every file in lattifai/lattifai-skills

Browse by kind, by grade A, or by owner.