Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

Lazycow Studio

MCPB · LAZYCOW-0.25.3.MCPB · 2 COMPONENTS

Drive Lazycow Studio by kensaurus: make reels and trip films from your footage, plan and post them.

45 Trust /100

Last scanned (17 hours ago)

Scores higher than 75% of MCPB servers

Category Marketing and SEO

Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →

Supply Chain Security13
  • No malware scan is available for this kind of package: the supply-chain vendors we use do not cover it. This is a permanent gap in our coverage, not a finding about the package.Unverified
  • Known CVEs could not be checked: we found no dependency list to check in this bundle: either it runs a compiled binary, whose bundled manifests we do not read, or none of the package.json, requirements.txt, pyproject.toml or uv.lock files it ships for its runtime declares a dependency we can read (a requirements.txt that only points at a setup.py project, a local path, or a nested requirements file we cannot follow counts as none).Unverified
  • No install/post-install scripts declared.Pass
  • Dependency health could not be checked: we found no dependency list to check in this bundle: either it runs a compiled binary, whose bundled manifests we do not read, or none of the package.json, requirements.txt, pyproject.toml or uv.lock files it ships for its runtime declares a dependency we can read (a requirements.txt that only points at a setup.py project, a local path, or a nested requirements file we cannot follow counts as none).Unverified
Provenance & Transparency48
  • Source repository is publicly reachable at the declared URL. View diagnostics → Pass
  • Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
  • Clear OSI-approved license (Apache-2.0).Pass
  • Actively maintained (last published 0 days ago).Pass
  • Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability71
  • AI-judged instruction clarity (good).Pass
  • Context-footprint check failed: tool/resource definitions use about 3893 tokens (~125/item across 31 items; 31 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management10
  • Stability observed for 3 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage97
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 91% of tool parameters carry a description.Partial
Tool Safety100
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.Pass
  • An AI judge read all 32 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a current MCP spec version (2026-07-28).Pass
What we tested

Checks we ran (13)

  • Install scripts
  • Source repository
  • Provenance
  • License
  • Maintenance
  • Security disclosure
  • Schema quality
  • Stability
  • Tool coverage
  • Injection markers
  • Destructive annotations
  • Judged manipulation
  • MCP protocol

Checks we could not complete (3)

These are gaps on our side, not findings about this package. We only credit what we can confirm, so a category made up entirely of them scores 0.

  • Malware scan in Supply Chain SecurityNo malware scan is available for this kind of package: the supply-chain vendors we use do not cover it. This is a permanent gap in our coverage, not a finding about the package.
  • Known CVEs in Supply Chain SecurityKnown CVEs could not be checked: we found no dependency list to check in this bundle: either it runs a compiled binary, whose bundled manifests we do not read, or none of the package.json, requirements.txt, pyproject.toml or uv.lock files it ships for its runtime declares a dependency we can read (a requirements.txt that only points at a setup.py project, a local path, or a nested requirements file we cannot follow counts as none).
  • Dependency health in Supply Chain SecurityDependency health could not be checked: we found no dependency list to check in this bundle: either it runs a compiled binary, whose bundled manifests we do not read, or none of the package.json, requirements.txt, pyproject.toml or uv.lock files it ships for its runtime declares a dependency we can read (a requirements.txt that only points at a setup.py project, a local path, or a nested requirements file we cannot follow counts as none).

Does not apply to this package (2)

  • Endpoint SecurityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
  • Transport & ReachabilityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.

The score does not measure everything. What the score does not measure →

Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 21 Nov 26 0
    • Provenance: fail → unverified ▼ security
    • Maintenance: pass → unverified ▼ functional
    • Package version: 0.24.1 → 0.25.3 functional
    • Package version: 0.24.1 → 0.25.1 functional
    • Package version: 0.24.1 → 0.25.0 functional
  • 20 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 3 to 7. That category is still filling its 30-day observation window: 1 days of observed history at the previous scan, 2 at this one. The score rises as the window fills, whether or not the server changes.

  • 19 Nov 26 0
    • Stability: unverified → 0.03 ▲ functional
    • Package version: 0.24.0 → 0.24.1 functional
  • 18 Nov 26 44

    First indexed and scored.

Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 21 Nov 2026 · Analysed mcpb/https://github.com/kensaurus/lazycow-studio/releases/download/v0.25.3/lazycow-0.25.3.mcpb@0.25.3

Provenance No attestation

The registry publishes no build provenance for this version, so there is nothing to verify.

Result No attestation
Ecosystem mcpb
Reason No attestation published

Background: How many MCP packages publish verified provenance →

MCP tools · 31 exposed · ~3,604 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
ai_estimate ~212

Price AI tools on one reel before anything is spent: review (a video model watches the reel and suggests fixes), sharpen (soft or low-res clips), slowmo (smooth slow motion), living (a real still gets a slow camera move), foley (natural sound made from the picture for shots that have none), behind (the headline sits behind the opening shot's subject), music (an instrumental bed made for the reel), voice (a Japanese AI voice reads the approved words). Returns the cost per tool, the total, the per-reel / total / daily limits and whether it fits. Spends nothing.

NameTypeReqDescription
intostring–Price it as an AI copy under this new slug (e.g. "day-1108-ai"); the original stays as it is.
slugstringyesThe reel, e.g. "day-1108".
toolsarrayyesSome of: review, sharpen, slowmo, living, foley, behind, music, voice.

No output schema declared.

No examples provided.

ai_review ~158

A video model watches the rendered reel (picture, sound, captions with their times) and returns a hook score for the first 3 seconds, a look score, captions that do not match their shot, on-screen claims that are not approved facts, and 1-3 concrete fixes. Suggestions only: nothing in the reel is changed. Costs about $0.01-0.03 (OpenRouter), inside the studio's AI limits; the same render is reviewed once and then answered free from its cache.

NameTypeReqDescription
freshboolean–Review again even if this render was already reviewed (costs again).
last_onlyboolean–Only return the last stored review; never spends.
slugstringyesA rendered reel.

No output schema declared.

No examples provided.

ai_run ~216

Apply AI tools to a reel as a background job, only inside the studio's AI limits: it prices the tools first and refuses (nothing spent) when the cost would pass the per-reel, total or daily limit. Tools never change an approved or scheduled reel: pass `into` to make an AI copy under a new slug. The result stays a draft for a person to review; nothing is approved or posted. Voice and living photo need the platform's AI label when posted. Poll job_status with the returned slug.

NameTypeReqDescription
intostring–A new slug for an AI copy (recommended), e.g. "day-1108-ai".
max_usdnumber–Refuse if the estimate is above this (on top of the studio's own limits).
renderboolean–Render the reel when the tools finish.
slugstringyesThe reel.
toolsarrayyesSome of: review, sharpen, slowmo, living, foley, behind, music, voice.

No output schema declared.

No examples provided.

cancel_job ~62

Stop a queued or running render or sound remix. A stopped render may leave half-made files: render the reel again before anyone approves it. Uploads, posts and footage scans cannot be stopped.

NameTypeReqDescription
slugstringyesThe reel slug of the render to stop.

No output schema declared.

No examples provided.

check_connections ~122

Before posting: checks each network in use. Does each sign-in still work (YouTube, Facebook, Instagram Stories, Bluesky)? Does it carry every permission the studio uses (YouTube scopes, Meta permissions such as instagram_manage_insights)? Is Chrome linked for TikTok, X, LinkedIn and the other networks your Chrome posts to? What was the latest post that failed in the last 14 days? Each problem comes with the exact fix and, where there is one, the page to open. Read-only: it never signs in or changes a connection.

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

job_status ~85

Progress of background jobs. With a slug: that job (a render or remix uses the reel slug; a post uses "post-<slug>-<network>"; a YouTube upload uses the reel slug; footage scans use "footage"). Without a slug: every render that is queued or running.

NameTypeReqDescription
slugstring–Job key; leave out to list active renders.

No output schema declared.

No examples provided.

list_accounts ~51

Which networks are switched on for this project and under which account name, which networks the studio can post to by itself and whether each is connected, and the YouTube connection. Never returns passwords, tokens or keys.

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

list_comments ~112

Comments on the project's YouTube videos, Bluesky posts and Facebook Page posts: who, what, when, on which reel, whether it is a question (and on what topic: price, booking, access, Wi-Fi...), praise or spam, and the reply drafted from the owner's saved answers. Read-only: it never replies; sending is a person's action in the studio. check=true reads the networks again first.

NameTypeReqDescription
checkboolean––
statestring–open = waiting for an answer.

No output schema declared.

No examples provided.

list_posts ~111

The posting log: every planned reminder, queued, scheduled, posted, failed or dropped post, newest first, with network, time and link. Filter by reel, network or status.

NameTypeReqDescription
limitinteger––
networkstring–Only this network, e.g. youtube, bluesky, facebook, instagram.
slugstring–Only posts of this reel.
statusstring–Only this status, e.g. planned, queued, scheduled, posted, failed, dropped.

No output schema declared.

No examples provided.

list_projects ~34

The studio's projects (one per brand or account: its own reels, posts and logins) and which one is open.

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

list_reels ~149

List the reels in the open project, newest first, with their review status (draft = not made yet, rendered = waiting for a person to review, approved = ready to post, rejected, reroll = asked to be remade), length, music, faces/consent and any render job in progress. Filter by status or theme.

NameTypeReqDescription
include_houseboolean–Also list the built-in house reels (view, fire, field, winter).
limitinteger––
statusstring–Only reels in this review status. "making" = a render is queued or running.
themestring–Only reels of this theme (see list_themes).

No output schema declared.

No examples provided.

list_themes ~43

The themes the studio can make a reel for (each is one story with a number of shot slots). Use a theme key with make_reels or make_reel.

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

make_reel ~113

Make one reel for a theme under a slug you choose, from the tagged footage, and queue it to render. Use make_reels to make one per theme automatically.

NameTypeReqDescription
facesboolean–Allow clips with faces.
sincestring–Use footage shot on or after this date (YYYY-MM-DD).
slugstringyesA new name: lowercase letters, digits and dashes, e.g. "fire-1108".
themestringyesTheme key from list_themes.

No output schema declared.

No examples provided.

make_reels ~191

Start making new reels: one per theme from the newest footage (clips picked by the studio, approved words, licence-clean music), each queued to render one after another. Returns the new slugs; follow each with job_status. Rendering takes minutes per reel. The new reels still need a person's review and approval in the studio before they can be published.

NameTypeReqDescription
cutstring–short = the looping reel; long = a 40-60 s explainer for LinkedIn and Facebook.
facesboolean–Allow clips with faces (they then need consent or Blur faces before approval).
sincestring–Use footage shot on or after this date (YYYY-MM-DD). Default: the last 60 days.
themesarray–Theme keys from list_themes. Leave out for every theme.
varyboolean–Give each reel a different text layout and effects.

No output schema declared.

No examples provided.

perf_posts ~73

Every post that went out in the last days (default 90), newest first, one row per network: views, likes, comments, shares, saves, average % watched and engaged views, from each post's latest reading; the best post of each network is marked. Read-only.

NameTypeReqDescription
daysinteger––

No output schema declared.

No examples provided.

perf_suggest ~145

Suggestions for the next reels: picture look, word recipe, text layout, length and posting hour (JST), each with why. Thompson sampling over what has worked, with a share of suggestions (default 30%) that explore less-tested choices on purpose; until 15 posts are measured every suggestion explores. Read-only: suggestions only steer how drafts are made; a person still approves every reel and nothing is posted.

NameTypeReqDescription
explorenumber–Share of suggestions that explore.
ninteger–How many suggestions.
networkstring–Suggest for one network. Leave out for all networks pooled.
seedinteger–Same seed, same suggestions.

No output schema declared.

No examples provided.

perf_summary ~133

What worked and what did not across posted reels, by choice (look, recipe, text layout, length, hook, music, theme, posting weekday and hour), in plain sentences with their uncertainty, e.g. "punch look: about 1.6x the median views on TikTok (5 posts; still uncertain)". Each post is compared with its own network's median. Read-only; below 15 measured posts it says the numbers are early hints.

NameTypeReqDescription
networkstring–One network (youtube, tiktok, facebook, bluesky, x, ...). Leave out for all networks pooled.

No output schema declared.

No examples provided.

plan_post ~137

Add a reminder to the posting log: post this reel on this network at this time. Nothing is posted; a person (or publish_reel later) does the posting. Times are in the audience's time zone unless an offset is given.

NameTypeReqDescription
atstringyesWhen: YYYY-MM-DDTHH:MM, optionally with +HH:MM, e.g. 2026-11-09T19:00+09:00.
networkstringyese.g. youtube, instagram, tiktok, facebook, bluesky, x, linkedin, threads.
slugstringyesThe reel's slug.

No output schema declared.

No examples provided.

publish_reel ~205

Post an APPROVED reel to a network now, or schedule it with `at`. Only reels a person has approved in the studio can be published: this tool checks the review status first and refuses anything that is not "approved". Networks the studio posts to itself: youtube (or youtube-shorts), and any connected network in list_accounts (e.g. bluesky, facebook). Runs as a background job; follow it with job_status.

NameTypeReqDescription
atstring–Schedule for this time (YYYY-MM-DDTHH:MM, optionally +HH:MM). Leave out to post now.
networkstringyesyoutube, bluesky, facebook, … (see list_accounts).
post_idinteger–Optional: the id of a planned post from list_posts that this post fulfils.
privacystring–YouTube only. Default private; a scheduled upload stays private until its time.
slugstringyesThe approved reel's slug.

No output schema declared.

No examples provided.

read_log ~101

The studio's activity log, newest first: finished jobs (renders, remixes, uploads, posts, footage scans, with how long they took and the last line on failure), paid AI spend, how each reel was assembled, or YouTube quota use.

NameTypeReqDescription
failed_onlyboolean–jobs only: just the jobs that ended in an error.
linesinteger–How many of the newest rows to return.
sectionstring––

No output schema declared.

No examples provided.

reel_details ~72

Everything about one reel: title, review status and notes, every shot (file, timing, role), music and sound, the files prepared for each network, and its current render job.

NameTypeReqDescription
slugstringyesThe reel's slug, e.g. "fire-1107" (from list_reels).

No output schema declared.

No examples provided.

remix_reel ~51

Re-render only the sound of a reel (music and effects mix), keeping the picture. Queued behind any running render; follow it with job_status.

NameTypeReqDescription
slugstringyesThe reel's slug.

No output schema declared.

No examples provided.

research_events ~96

Find upcoming events near the open project's home place (Settings › Keys & AI › This project) in the next N days, with their dates and a suggested post-by date, as ideas for when to post. Dates come from search text: confirm on the event page. Read-only.

NameTypeReqDescription
daysinteger–How many days ahead to look.
freshboolean–Skip the 24 h cache (spends credits again).

No output schema declared.

No examples provided.

research_ideas ~140

Search recent web pages and news (Japanese and English) about a topic (default: the open project's name and home place) and return the pages with their URLs, frequent hashtags and keywords, and 3-5 suggested post angles, each citing the pages it came from. Read-only: suggestions for a person, never post copy, never posted. Uses Firecrawl credits (cached 24 h).

NameTypeReqDescription
daysinteger–How far back to look.
freshboolean–Skip the 24 h cache (spends credits again).
topicstring–A few words to search for. Leave out for the project's default topic.

No output schema declared.

No examples provided.

research_links ~76

Check that every link in planned or scheduled posts (their post text and live link) and in the profile kit still loads. Reports broken links with where each is used; sites that refuse automatic checks are listed as unsure. Read-only.

NameTypeReqDescription
freshboolean–Skip the 24 h cache for the few links checked through Firecrawl.

No output schema declared.

No examples provided.

research_page ~110

Read one public web page (a competitor, an inspiration) and return its summary, key selling points, amenities, prices it states, and what it says that the studio's approved facts do not. Read-only; their words are for ideas, never to be copied into a post. X and LinkedIn pages are refused (Firecrawl bills them through paid providers).

NameTypeReqDescription
freshboolean–Skip the 24 h cache (spends credits again).
urlstringyesThe page's address.

No output schema declared.

No examples provided.

research_site_check ~135

Read the brand website (the project's profile link and its booking page) and compare its prices, key facts, inclusions and booking link with the facts the studio's approved copy uses. Returns each fact as match, mismatch, not on the site or only on the site, in plain words (e.g. "the website says ¥5,500; your copy says ¥5,200"). Read-only: it changes no copy and posts nothing.

NameTypeReqDescription
freshboolean–Skip the 24 h cache (spends credits again).
urlstring–Another page to check instead of the brand website (optional).

No output schema declared.

No examples provided.

schedule_reel ~269

Schedule an APPROVED reel on X, TikTok or Google Maps (gbp) through the studio's browser helper, which has no API: it attaches the reel's file, writes its approved text, sets the date and time, checks what the site shows, presses Schedule, and records the post as scheduled in the posting log with a proof link. Refuses any reel whose review status is not "approved", a second copy of the same reel on the same network at the same time, and times outside the site's window. Runs as a background helper job, one at a time; follow it with job_status slug=helper-<slug>-<network>. If the helper window is not signed in to the site, the job ends asking the owner to sign in once (Accounts in the studio).

NameTypeReqDescription
atstringyesWhen, with its zone: e.g. 2026-11-16T19:00+09:00 (19:00 JST). The helper converts it to the browser's clock.
networkstringyesx, tiktok or gbp (Google Maps / Business Profile).
rehearseboolean–Do every step but the final press (nothing is scheduled or logged).
slugstringyesThe approved reel's slug.

No output schema declared.

No examples provided.

studio_status ~54

Check that Lazycow Studio is running and healthy: version, which project is open, whether this computer can render (with a fix for each problem), and what the render queue is doing. Call this first if anything else fails.

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

switch_project ~63

Open another project. The studio restarts on that project's folders (a few seconds); it refuses while any job is still running. Every later tool call works on the new project.

NameTypeReqDescription
idstringyesProject id from list_projects ("main" is the first one).

No output schema declared.

No examples provided.

youtube_channel ~85

Each connected YouTube channel: subscribers and total views, Shorts / long videos / live summed separately, and the newest uploads (including ones the studio did not make) with views, engaged views, likes, comments, watch time and % watched. Cached 3 hours; fresh=true reads YouTube again (about 10 quota units). Read-only.

NameTypeReqDescription
freshboolean––

No output schema declared.

No examples provided.

Install

How do I install the Lazycow Studio MCP server?

Lazycow Studio ships as an MCPB bundle, a single file you download and open in an MCP host that supports MCPB bundles, such as Claude Desktop. The host reads the launch command from the bundle's own manifest, so there is no command to copy. Before opening it, check the file's SHA-256 matches the one the MCP registry declares for it, 7a1a02bc3d39f00cd094db1fff3e724e0d6b6651203e83920e98f5c4cdb37ad0; the commands to do that are on this page. If it does not match, do not open it.

mcpb · lazycow-0.25.3.mcpb

Download bundle

Check the download matches the SHA-256 the MCP registry declares before you open it. If it does not match, do not open the file.

SHA-256

7a1a02bc3d39f00cd094db1fff3e724e0d6b6651203e83920e98f5c4cdb37ad0

macOS / Linux

echo "7a1a02bc3d39f00cd094db1fff3e724e0d6b6651203e83920e98f5c4cdb37ad0  lazycow-0.25.3.mcpb" | shasum -a 256 -c

Windows (PowerShell)

(Get-FileHash -Algorithm SHA256 .\lazycow-0.25.3.mcpb).Hash -eq "7a1a02bc3d39f00cd094db1fff3e724e0d6b6651203e83920e98f5c4cdb37ad0"

Run it in the folder you saved the bundle to, then open the file with an MCPB-capable host such as Claude Desktop.

Common questions

What is the Lazycow Studio MCP server?

Lazycow Studio is an MCP server listed in the public MCP registry as io.github.kensaurus/lazycow. Drive Lazycow Studio by kensaurus: make reels and trip films from your footage, plan and post them. This page covers its MCPB bundle (https://github.com/kensaurus/lazycow-studio/releases/download/v0.25.3/lazycow-0.25.3.mcpb).

Is the Lazycow Studio MCP server safe to use?

Lazycow Studio scores 45 out of 100 on VerifyMCP. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the Lazycow Studio MCP server expose?

Lazycow Studio exposes 31 tools: studio_status, list_reels, reel_details, list_themes, make_reels, and 26 more. Their descriptions and schemas cost roughly 3,604 tokens of context every time the server is loaded.

What licence is the Lazycow Studio MCP server under?

Lazycow Studio declares the Apache-2.0 licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

What did VerifyMCP test on the Lazycow Studio MCP server?

VerifyMCP ran 13 checks on Lazycow Studio, covering Install scripts, Source repository, Provenance, License and 9 more. 3 items could not be completed, a gap on our side rather than a finding about Lazycow Studio: Malware scan, Known CVEs and Dependency health. Endpoint Security and Transport & Reachability do not apply to an MCPB bundle, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.