Onshape (BYOK)
Browse your Onshape CAD documents, versions, elements, parts, mass properties and assembly bills of materials with your own Onshape API access key and secret
Caption/subtitle duration, character-per-line and reading-pace calculator, sourced from real style guides
Unify Dynamics LLC
Real questions, and the tools your agent calls to answer them.
recommended_durationcharacters_per_line_checksplit_into_linespacing_checkestimate_script_caption_countconvert_reading_speedcaption_profile_referencecaption_glossaryCapshape is a fully local caption/subtitle timing & readability calculator. recommended_duration computes how long a caption should stay on screen from its character count and a reading-speed profile, clamped to that profile's min/max duration. characters_per_line_check checks each line of a caption against a profile's max characters-per-line and max-lines limits. split_into_lines word-wraps a longer caption into balanced lines under a character-per-line and max-lines limit. pacing_check takes a caption and its assigned duration and verdicts whether the pace is too fast, too slow, comfortably loose, or well-paced. estimate_script_caption_count estimates how many caption cards a full script will need and the total on-screen caption time. convert_reading_speed converts between words-per-minute and characters-per-second. caption_profile_reference gives full detail for a named profile (netflix_adult: 20 CPS/42 CPL; netflix_children: 17 CPS/42 CPL; bbc_general: ~17 CPS/37 CPL) sourced from Netflix's Partner Help Center and BBC subtitling guidance, or lists them all. caption_glossary defines CPS, CPL, lead-in/lead-out time, gap, shot-change sensitivity, SDH, closed vs burned-in captions, reading speed. Keyless; fully local; zero upstream calls; nothing stored.
Sources: Real computation (duration/CPS/line-wrap arithmetic) against reading-speed and layout profiles sourced from Netflix's Partner Help Center Timed Text Style Guide and BBC subtitling guidance, verified 2026-09-25; fully local, zero upstream calls
The publisher, Unify Dynamics LLC, states that tool inputs are processed in memory and not retained.
8 tools, grouped by job. Every one is read-only or compute-only.
https://capshape.connectorbay.com/mcpBrowse your Onshape CAD documents, versions, elements, parts, mass properties and assembly bills of materials with your own Onshape API access key and secret
Read your Wowza Video live streams, transcoders, recordings, stream targets, players, schedules and usage with your own access token
Read your Twelve Labs video understanding indexes, indexed videos, assets and entity collections with your own API key
Read your Hedra wallet balance, usage, transactions, model catalog, voices and generation job status with your own API key
Read your getimg.ai model catalog, API balance, daily costs and video generation status with your own API key
Read your Duda website builder sites, pages, blog posts, templates, collections and content library with your own API user and password
"How long should this caption stay on screen?", "does this line fit Netflix's character limit?", "is 3 seconds too fast for this caption?" — real reading-speed math against the same profiles professional subtitlers actually work to, not a guess.
recommended_duration computes how long a caption should stay on screen from its character count and a
reading-speed profile, clamped to that profile's minimum and maximum duration. characters_per_line_check checks
each line of a caption against a profile's character-per-line and max-lines limits. split_into_lines word-wraps
a longer caption into balanced lines that fit those limits. pacing_check takes a caption and the duration
you've actually assigned it and verdicts whether it's too fast, too slow, comfortably loose, or well-paced.
estimate_script_caption_count estimates how many caption cards a full script or transcript will need and the
total on-screen caption time. convert_reading_speed converts between words-per-minute and characters-per-second.
caption_profile_reference gives full detail for a named profile — netflix_adult (20 CPS, 42 characters per
line), netflix_children (17 CPS), bbc_general (~17 CPS, 37 characters per line, converted from the BBC's
published 160-180 words-per-minute guidance) — or lists them all. caption_glossary defines core terms: CPS,
CPL, lead-in/lead-out time, gap, shot-change sensitivity, SDH, closed captions, burned-in captions, reading speed.
Not a captioning tool itself — it computes timing and layout numbers, it doesn't generate or burn in captions. Profile numbers are sourced from Netflix's Partner Help Center Timed Text Style Guide and BBC subtitling guidance as published at the connector's build date; style guides are occasionally revised, so re-check the original source before treating a number here as a hard delivery requirement.
Effective 25 September 2026. Unify Dynamics LLC ("we", "us"), Menlo Park, California.
What Capshape is. Capshape is a connector (an MCP server and REST API) that AI assistants and personal agents call on your behalf for a fully local caption/subtitle timing and readability calculator, computing recommended duration, character-per-line checks, line-splitting and reading pace against profiles sourced from published subtitle style guides. You interact with your assistant; the assistant sends us only the arguments needed for that one lookup or calculation.
What we receive. The arguments of each tool call — caption text, a duration, a reading-speed profile name, or a conversion value — the technical headers needed to serve the request (an IP address or a client identifier supplied by the assistant, and a Pro API key if you have one), and standard server logs.
Everything is answered from bundled reference data and/or local computation, and nothing is stored. No request is forwarded to any third party — there is no upstream call at all, live or otherwise — and no query is logged in request bodies, recorded, or used to train anything.
What we do not receive or store. We do not receive your conversation with the assistant or your name, and we do not build a profile from your queries: each request is answered independently and nothing carries over between requests. Rate-limit counters (a client identifier and timestamps) are held in memory for at most 24 hours. Hosting-provider access logs, which contain IP addresses and request paths but not request bodies, are retained for up to 30 days for security and abuse prevention.
Sub-processors. None — the entire dataset is bundled with the server and answered locally. Hosting: Vercel (United States regions).
Legal bases and your rights. Our processing is necessary to provide the service you request through your assistant. Because we store no personal data beyond short-lived logs, requests to access, correct or delete personal data are usually satisfied automatically; you can still contact agent@connectorbay.com and we will respond within 30 days. We do not sell or share personal information and do not use your data to train models.
Children. Capshape is a general-audience service operated by the adult using the assistant.
Changes. We will post changes here with a new effective date.
Effective 25 September 2026. These terms govern use of the Capshape connector ("the Service") operated by Unify Dynamics LLC ("we", "us").
Email agent@connectorbay.com. We answer within two business days.
https://capshape.connectorbay.com/mcphttps://capshape.connectorbay.com/openapi.jsonWhen you report a problem, include the question you asked your agent, roughly when, and which agent you used (for example Claude, ChatGPT or Gemini). Never send us passwords, account numbers or ID numbers. Caption Timing never asks for them.