All tools

Every one of the 67 tools, with its complete option list. These pages are generated from cleya capabilities at docs build time, so they cannot drift from the toolkit: add a flag, rebuild, the page is right.

Generated from cleya v0.6.0.

perception — instruments — measure what was rendered (WCAG contrast, alignment, brand conformance, copy claims, pixel diff)

Tool
`align`	Optical-alignment check
`brand_check`	Brand-compliance score
`claims`	Lint a creative's copy against a claims policy
`contrast`	WCAG contrast check on rendered text
`diff`	Compare two rendered frames
`pacing`	Measure a rendered film's rhythm
`palette`	Color report for a frame
`review`	Perception QA on a rendered video
`scrub`	Open a film in the director's chair

author — create media from specs (scene DSL, cards, titles, captions, voiceover, capture)

Tool
`animated-captions`	Add karaoke-style animated captions (macOS only, requires Whisper)
`autosub`	Auto-generate and burn subtitles (requires Whisper)
`caption`	Burn captions/subtitles into video
`capture`	Record a real page driven through a scripted sequence to video (product demos, walkthroughs)
`card`	Generate a video card with text (intro/outro)
`compose`	Assemble scenes into one video
`gif`	Export video segment as optimized GIF
`kenburns`	Create Ken Burns effect (zoom/pan animation) from a still image
`render`	Render an animation to video
`screenshot`	Render an HTML file or URL to a PNG (single shot, lossless
`still`	Turn a still image into a video clip
`title`	Burn styled text titles into video (location, lower third, chapter, date)
`tts`	Generate a voiceover clip from text via the OpenAI audio API (opt-in cloud; needs OPENAI_API_KEY)

edit — transform existing media

Tool
`audio`	Extract, replace, remove, or mix audio
`bg-remove`	Remove video background using AI person segmentation (macOS only)
`clean-speech`	Detect and remove filler words (um, uh, you know, etc
`color`	Apply color grading preset
`concat`	Concatenate video clips with optional transitions
`draft`	Instant draft
`edit`	Deterministic scene edits
`effects`	Apply cinematic visual effects (grain, vignette, bloom, filmstrip)
`harvest`	Promote an approved scene to a reusable starter (the harvest loop)
`layout`	Arrange multiple videos in a grid or split-screen
`lut`	Apply a
`multiclip`	Switch between camera angles using a cut list
`overlay`	Add image overlay (logo, watermark)
`pip`	Add picture-in-picture overlay
`reformat`	Reformat video for a platform (TikTok, Reels, Shorts, etc
`shake`	Add camera shake effect (subtle, moderate, heavy)
`speed`	Adjust playback speed (0
`sync`	Align multiple camera recordings by audio cross-correlation
`transition-overlay`	Apply a video transition clip between two clips
`trim`	Trim video to a time range
`upscale`	Upscale video resolution with high-quality interpolation

scout — understand the material before acting

Tool
`beat-detect`	Detect beats and estimate BPM from audio
`caption-extract`	Extract caption styling from a reference video
`cut-silence`	Detect silent segments in a video
`detect-text`	Detect text (OCR) in video frames or a loose image (macOS only)
`extract-candidates`	Find and rank clip-worthy segments in a video
`image-similarity`	Cluster near-duplicate / same-shoot images by Vision feature-print (macOS only)
`info`	Get video metadata (duration, resolution, codec, fps, bitrate)
`loudness`	Analyze audio loudness over time (LUFS, peaks, timeline)
`lut-extract`	Extract color LUT from a reference video
`motion`	Score visual motion/activity across a video
`scenes`	Detect scene cuts in a video
`starters`	Starter scene library
`thumbnail`	Extract a frame as an image
`thumbnails`	Extract multiple thumbnails in one call (batch mode)
`transcribe`	Transcribe audio to SRT subtitle file (requires Whisper)
`virality-score`	Score a clip 0-100 for predicted social media engagement

assets — extract usable pieces from media

Tool
`brand_extract`	Reference image → starter BrandProfile JSON: bg/ink/accent/muted roles assigned from dominant colors (deterministic; edit before trusting)
`cutout`	Lift ANY salient subject (product/object/animal/person) onto transparency
`segment`	Detect people and generate segmentation mask (macOS only)
`smart_crop`	Crop to a target aspect while keeping the subject

export — hand off to other tools (NLE timelines, translated subs)

Tool
`export-nle`	Export edit plan as FCPXML for Premiere Pro / DaVinci Resolve
`translate-srt`	Translate subtitle files to other languages

system — utilities

Tool
`list-transitions`	List available transition clips in a directory

pipeline — multi-step flows in one call

Tool
`social-clip`	Generate platform-ready clips from best segments

perception — instruments — measure what was rendered (WCAG contrast, alignment, brand conformance, copy claims, pixel diff)​

author — create media from specs (scene DSL, cards, titles, captions, voiceover, capture)​

edit — transform existing media​

scout — understand the material before acting​

assets — extract usable pieces from media​

export — hand off to other tools (NLE timelines, translated subs)​

system — utilities​

pipeline — multi-step flows in one call​