Skip to content
Official MCP RegistryListed

VOOPA Render Check

Part ofVOOPA Render CheckMCP server

Review AI images, videos and audio against a brief before paying to rerender.

First seen 6 Nov 2026. Evidence as of 18 Nov 2026.

16
Tools
From an anonymous probe
1
Source listings
Each with its own history
28
Recorded changes
Since first seen

Tools

ToolDescriptionBehaviour
check_black_gapsFlag nearly black frames in a selected 1–120 second range for manual review. The default minimum is 0.3 seconds; minimumDurationSeconds can be 0.1–3 seconds. Fades, dark scenes and black frames may be intentional; a clear result does not verify the edit. Timestamps refer to the original video.Read-only
check_full_frame_freezeFind possible freezes in a selected 1–120 second video range or a caller-selected fixed region. Set minimumFreezeSeconds from 0.25 to 5 (default 1); shorter thresholds may flag intentional stills. For longer videos the default is the first 120 seconds; pass fromSeconds and durationSeconds to inspect another section. Timestamps refer to the original video. Flags require human review. A fixed rectangle does not track objects; camera movement and intentional stills confound results. Does not certify natural motion.Read-only
check_image_exportCheck a PNG, JPEG or WebP attachment or public HTTPS image URL against supplied orientation or minimum pixel dimensions. Encoded pixel dimensions may differ from EXIF display orientation. Does not assess sharpness, anatomy, text, content, or creative quality; resizing cannot create missing detail. Review before another paid render.Read-only
check_repeated_sequenceCompare two non-overlapping ranges of the same exported video for visual similarity. Returns a candidate and time ranges for human review, not a creative verdict. Does not verify audio, intent or all repetition; never authorizes spending.Read-only
check_video_exportCheck a completed video against explicitly supplied delivery requirements: orientation, duration and/or minimum display dimensions. Accepts a public HTTPS video URL or attachment; local hosts also accept an intake filename. Container metadata can be wrong and cannot verify creative content or image quality. Suggest trimming, reframing or re-export where appropriate; never approve spending.Read-only
compare_lyrics_transcriptFree word-by-word comparison of approved lyrics and a transcript of the matching sung or spoken excerpt, supplied by a reviewer or a transcription tool. Reports changed, omitted and extra words. Ignores punctuation and case but preserves accents; Japanese and other languages use word segmentation. This tool does not transcribe or hear audio, check lip sync or judge singing quality. Transcription of singing can be wrong; listen to differences and mark transcriptVerified only if independently checked. Empty transcripts are inconclusive. Do not use whole-song lyrics against a short excerpt.Read-only
compare_reference_audioCompare up to 15 seconds of a video soundtrack with a supplied corresponding audio excerpt. Set videoFromSeconds and referenceFromSeconds when the sections start at different times. Returns overall and one-second waveform-similarity signals for human review, not a verdict; quiet windows have no score. A low score can reflect mixing or a different excerpt; a high score does not prove lip sync, lyrics or intended sound.Read-only
compare_shot_inventoryFree deterministic check for shot IDs missing from a supplied inventory, unexpected or repeated IDs, and order. If all items have start/end times, also reports gaps and overlaps within the supplied single-track timeline. List delivered items in timeline order. Set inventoryComplete true only if the caller knows the inventory covers the whole edit. This does not inspect video, infer a shot from a prompt, identify an object or decide whether repeats, gaps or transitions are intentional. Review before another render.Read-only
decide_render_retryUse after inspecting a completed render. Supply one observation per requirement, past attempts in chronological order, and candidate fixes grounded in available assets or a specific untried change. Returns keep, edit first, propose one retry, change approach, or review. This tool never authorizes spend.Read-only
get_credit_optionsGet current usage prices and the customer account page. This does not create an order, access payment details or authorize spending.Read-only
inspect_audio_levelsMeasure decoded sample levels in a selected 1–30 second section of one video, optionally comparing another video. Reports RMS, peaks, near-full-scale sample counts and one-second windows for listening review. Does not identify a voice, prove clipping, measure perceptual loudness or verify lip sync. Video audio includes music and effects. An interval may be shorter than requested; check examinedSeconds. Never approves a paid render.Read-only
measure_region_motionProduce one-second pixel-change measurements in a caller-selected fixed rectangle over 1–30 seconds, for manual review of suspected still backgrounds or other motion. Camera movement, cuts, lighting and occlusion can create or conceal change; this is not object tracking or a verdict about natural motion. No result approves a paid retry.Read-only
prepare_continuity_comparisonExtract one labeled frame from each of two completed videos for a vision-capable agent or person to compare a specified continuity concern, such as a character, object position or costume. Accepts public HTTPS videos or attachments. Provides still evidence, not an automated continuity or motion verdict; differences may be intentional. Never approves spending.Read-only
review_completed_imageUse when a completed generated image is wrong or the user wants a redo. Inspect a PNG, JPEG or WebP attachment or direct public HTTPS image URL from any renderer, optionally with a reference image. Use the original brief and prior attempts when available. Compare the returned visual evidence and call decide_render_retry. This tool never starts a render.Read-only
review_completed_renderUse before paying to rerender. Review an exported MP4, MOV or WebM from any renderer. Use a chat attachment or a direct public HTTPS video URL. Include the brief, prior attempts, and any interval expected to have continuous audio. Returns timestamped evidence or an audio hold and lists checks needing manual review. Sampled frames do not verify lip-sync, soundtrack identity, continuous motion or the entire video. Uses host vision when supported.Read-only
scan_repeated_shotsExperimental visual scan of a selected 4–60 second range. For longer videos defaults to the first 60 seconds; use fromSeconds and durationSeconds to scan another section. Suggests similar two-second ranges for manual review, separating low-motion matches. Timestamps are relative to the original video. May miss crops, edits and repeats, and can flag intentional or similar footage. Audio and intent are not checked; an empty result does not approve the video.Read-only

Change history

  1. scan_repeated_shots: tool added
  2. review_completed_render: input schema changed
  3. review_completed_render: description changed (+"hold and lists checks needing manual review. Sampled frames do not verify lip-sync, soundtrack identity, continuous motion or the entire video." -"hold.")
  4. prepare_continuity_comparison: tool added
  5. measure_region_motion: tool added
  6. inspect_audio_levels: tool added
  7. get_credit_options: tool added
  8. compare_shot_inventory: tool added
  9. compare_reference_audio: description changed (+"overall and one-second" +"signals" +"verdict; quiet windows have no score.")
  10. compare_lyrics_transcript: tool added
  11. check_video_export: tool added
  12. check_image_export: tool added
  13. check_full_frame_freeze: input schema changed (+durationSeconds, +fromSeconds, +minimumFreezeSeconds)
  14. check_full_frame_freeze: description changed (+"selected 1–120 second" +"range" +"region. Set minimumFreezeSeconds from 0.25 to 5 (default 1); shorter thresholds may flag intentional stills. For longer videos the default is the first")
  15. check_black_gaps: tool added
  16. server instructions changed (+"For suspected repeated shots, offer scan_repeated_shots as an experimental candidate search and show the scanned range and limitations.")
  17. compare_reference_audio: tool added
  18. check_repeated_sequence: tool added
  19. check_full_frame_freeze: tool added
  20. Description changed (registry)
  21. review_completed_render: input schema changed (+videoAttachment)
  22. review_completed_render: description changed (+"an exported MP4, MOV or WebM from any renderer. Use" +"chat attachment" +"a direct public HTTPS")
  23. review_completed_image: tool added
  24. server instructions changed (+"When" +"completed" +"image or")
  25. review_completed_render: input schema changed (+expectedAudio, +renderVideoUrl)
  26. review_completed_render: description changed (+"before paying to rerender. Review" +"completed fal, Higgsfield, Runway, or Replicate" +"URL, or")
  27. Description changed (registry)
  28. Listed (registry)
Source listings
SourceListingFirst seenLast seenVersions
Official MCP Registryio.github.Voopa-Corp/render-check6 Nov 202618 Nov 20263
VOOPA Render Check on Official MCP Registry | InvokeRank