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.

Kinocut

PYPI · KINOCUT

Guardrailed video editing for AI agents: FFmpeg, captions, effects, Hyperframes, and receipts.

Available components

−15 this week 77 Trust /100

Last scanned (13 hours ago)

Scores higher than 76% of PyPI servers

Category Design and Media

Deprecated

This server is marked deprecated in the MCP registry. The registry records this reason: Moved to tech.kyanitelabs/kinocut, releases from 1.16.2 onward publish there (latest 1.16.5). Please switch server references.

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 Security50
  • Malware scan not yet available for this package.Unverified
  • No known CVEs affecting this package version or its production dependencies.Pass
  • Runs hatchling.build at install time, a recognised build step with no custom scripting around it. View diagnostics → Pass
  • 1 of 35 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency100
  • Source repository is publicly reachable at the declared URL. View diagnostics → Pass
  • Cryptographically verified build provenance (signed, bound to KyaniteLabs/kinocut). View diagnostics → Pass
  • Clear OSI-approved license (Apache-2.0).Pass
  • Actively maintained (last published 8 days ago).Pass
  • Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability77
  • 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
  • AI-judged instruction clarity (good).Pass
  • Context-footprint check failed: tool/resource definitions use about 22308 tokens (~109/item across 204 items; 203 tools + 1 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 Management100
  • No destabilizing schema changes in the last 30 days.Pass
Tool Coverage73
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 6% of tool parameters carry a description.Partial
  • Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety78
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • 1 of 10 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "video_acceptance_eval" implies "eval" and declares no destructiveHint at all, which the MCP spec reads as destructive by default. See how to fix → Partial
  • An AI judge read all 205 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
What we tested

Checks we ran (15)

  • Known CVEs
  • Install scripts
  • Dependency health
  • 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 (1)

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 SecurityMalware scan not yet available for this package.

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.

  • 5 Nov 26 −15
    • Malware scan: pass → unverified ▼ security
  • 2 Nov 26 0
    • Stability: 0.97 → pass security
  • 1 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 93 to 97. That category is still filling its 30-day observation window: 28 days of observed history at the previous scan, 29 at this one. The score rises as the window fills, whether or not the server changes.

  • 31 Oct 26 +14
    • Malware scan: unverified → pass ▲ security
    • Stability: pass → 0.93 functional
  • 30 Oct 26 −15
    • Malware scan: pass → unverified ▼ security
  • 28 Oct 26 +15
    • Malware scan: unverified → pass ▲ security
  • 27 Oct 26 0
    • Stability: fail → pass ▲ security
  • 18 Oct 26 0
    • Tool safety: pass → unverified ▼ security
    • Stability: fail → unverified ▼ security
    • Tool coverage: 100 → unverified ▼ functional
    • Capabilities: pass → unverified ▼ functional
    • Schema quality: 100 → unverified ▼ functional
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 5 Nov 2026 · Analysed pypi/kinocut@1.16.1

Provenance Verified

A signed build attestation was found and verified, binding this exact artifact to the source repository it claims to come from.

Result Verified
Ecosystem pypi
Reason Verified
Discovered via Registry attestation endpoint
Source repo KyaniteLabs/kinocut
Certificate issuer https://token.actions.githubusercontent.com
Certificate SAN https://github.com/KyaniteLabs/kinocut/.github/workflows/publish.yml@refs/tags/v1.16.1
Rekor log index 3103405812
Predicate type PyPI publish attestation https://docs.pypi.org/attestations/publish/v1
Subject digest sha256:f00db0290f7217f650ffea23edaf1c4b43c0121b320cc24f30e12442e00c2e9a

Background: How many MCP packages publish verified provenance →

Install scripts 1 script
Hook Tier Command
build_backend Known build tool hatchling.build

Background: Why install scripts are a supply-chain risk →

Dependencies 35 packages
Packages resolved 35
Stale 1
Tree resolution Complete

Background: SBOMs and build attestations, explained →

MCP tools · 203 exposed · ~22,219 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
video_workflow_render ~678

Execute an agent workflow job-spec and return a provenance receipt. Validates the spec first (fail-closed), then runs each allowlisted op (probe|trim|resize|convert|merge|add_text|composite_layers) SEQUENTIALLY in spec order via the backing engine functions. Intermediates are written to a per-run ``@work`` directory unique to this invocation and cleaned on success (kept on failure); final media lands at the declared ``@outputs`` paths. Batch variants: pass ``variant=<id>`` to render one declared variant (its overrides patch the shared steps/outputs, and the single ``@outputs`` path is auto-named with the variant id so N variants emit N distinct outputs); the receipt records ``workflow.variant``. Pass ``all_variants=True`` to render EVERY declared variant in turn and return a ``workflow_batch`` summary (one receipt per variant, each into its own ``@work`` dir); use ``save_receipt_dir`` to also write each variant's receipt to ``<dir>/<variant>.json``. ``variant`` and ``all_variants`` are mutually exclusive. Pass ``keep_intermediates=True`` to retain ``@work`` intermediates even on success (recorded as the ``keep-intermediates`` cleanup policy). Pass ``resume_receipt`` (a prior render receipt from a job that failed with its intermediates kept) to RESUME: the current spec_hash must equal the receipt's (else fail-closed ``resume_spec_mismatch``) AND, for a variant, the receipt's variant must match (else ``resume_variant_mismatch``); each step whose recorded status is ``completed`` AND whose recorded input hashes still match AND whose recorded output file still exists and re-hashes to the recorded hash is SKIPPED, and the first step failing any check plus everything after it re-runs. Returns a workflow receipt (``receipt_kind: "workflow"``) capturing tool + FFmpeg versions, the spec hash, per-source probes/hashes, per-step status with real sha256 hashes of every consumed input and produced output, the cleanup manifest, and the determinism-scope caveat. On the first f…

NameTypeReqDescription
all_variantsboolean––
keep_intermediatesboolean––
resume_receipt–––
save_receipt–––
save_receipt_dir–––
spec_pathstringyes–
variant–––

Structured output declared, but exposes no named fields.

No examples provided.

video_workflow_validate ~226

Validate an agent workflow job-spec without rendering any media. Runs the fail-closed structural validator over the JSON job-spec at ``spec_path``: op allowlist (probe|trim|resize|convert|merge|add_text|composite_layers), symbolic ``@ref`` resolution (@sources.<id>, @work/<name>, @outputs.<id>), backward-reference-only ordering (a step may reference @work outputs from strictly-earlier steps only), per-op param introspection, and workspace-confined path safety (absolute paths and ../ / symlink escapes fail closed). Returns a structured verdict (``{"valid": true, ...}``) on success. On any structural violation it fails closed with a specific error ``code`` (``invalid_workflow_spec``, ``unknown_workflow_ref``, ``unsupported_workflow_op``, ``unsafe_workflow_source``, ``invalid_workflow_params``). Args: spec_path: Absolute path to the workflow job-spec JSON file.

NameTypeReqDescription
spec_pathstringyes–

Structured output declared, but exposes no named fields.

No examples provided.

video_write_metadata ~94

Write metadata tags to a video/audio file. Args: input_path: Absolute path to the input file. metadata: Dict of tag key-value pairs (e.g. {'title': 'My Video', 'artist': 'Me'}). output_path: Where to save the output. Auto-generated if omitted.

NameTypeReqDescription
input_pathstringyes–
metadataobjectyes–
output_path–––

Structured output declared, but exposes no named fields.

No examples provided.

Install

How do I install the Kinocut MCP server?

Kinocut runs locally as a PyPI package, launched with uvx kinocut. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

pypi · kinocut

# add to Claude Code
claude mcp add kyanitelabs-kinocut -- uvx kinocut
Common questions

What is the Kinocut MCP server?

Kinocut is an MCP server listed in the public MCP registry as io.github.KyaniteLabs/kinocut. Guardrailed video editing for AI agents: FFmpeg, captions, effects, Hyperframes, and receipts. This page covers its PyPI package (kinocut).

Is the Kinocut MCP server safe to use?

Kinocut scores 77 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 5 November 2026. Its build provenance is signed and verified. 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 Kinocut MCP server expose?

Kinocut exposes 203 tools: video_info, video_trim, video_merge, video_add_text, video_add_texts, and 198 more. Their descriptions and schemas cost roughly 22,219 tokens of context every time the server is loaded.

Is the Kinocut MCP server still maintained?

Kinocut is marked deprecated in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

What licence is the Kinocut MCP server under?

Kinocut 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 Kinocut MCP server?

VerifyMCP ran 15 checks on Kinocut, covering Known CVEs, Install scripts, Dependency health, Source repository and 11 more. One item could not be completed, a gap on our side rather than a finding about Kinocut: Malware scan. Endpoint Security and Transport & Reachability do not apply to a PyPI package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.