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.

Audio Studio

PYPI · AUDIO-STUDIO-MCP

Local AI BGM and sound effects with Stable Audio 3 (Apple Silicon MLX, NVIDIA CUDA)

Available components

71 Trust /100

Last scanned (23 minutes ago)

Scores higher than 64% of PyPI servers

Category Design and Media

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 Security82
  • No malware found by supply-chain analysis.Pass
  • CVE check failed: a known high-severity CVE affects cryptography 48.0.1, a direct dependency. A fixed version is available. View diagnostics → Fail
  • Runs hatchling.build at install time, a recognised build step with no custom scripting around it. View diagnostics → Pass
  • 5 of 87 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency97
  • Source repository is publicly reachable at the declared URL. View diagnostics → Pass
  • Cryptographically verified build provenance (signed, bound to kanaworksai-hue/audio-studio-mcp). View diagnostics → Pass
  • Clear OSI-approved license (MIT).Pass
  • Actively maintained (last published 5 days ago).Pass
  • Disclosure check failed: no security disclosure policy was found in the source repository. See how to fix → Fail
Schema Quality & AI Usability54
  • AI-judged instruction clarity (good).Pass
  • Context-footprint check failed: tool/resource definitions use about 2160 tokens (~540/item across 4 items; 4 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 Management0
  • Stability not yet verified: not enough scan history yet (needs a 30-day window).Unverified
Tool Coverage96
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 88% 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
  • We read all 4 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
  • An AI judge read all 5 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

Unverified: 1 category

A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.

What we tested

Checks we ran (15)

  • Malware scan
  • Known CVEs
  • Install scripts
  • Dependency health
  • Source repository
  • Provenance
  • License
  • Maintenance
  • Security disclosure
  • Schema quality
  • 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.

  • Stability in Stability & Change Management (scores 0 for this)Stability not yet verified: not enough scan history yet (needs a 30-day window).

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.

  • 16 Nov 26 71

    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 22 Nov 2026 · Analysed pypi/audio-studio-mcp@1.2.0

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 kanaworksai-hue/audio-studio-mcp
Certificate issuer https://token.actions.githubusercontent.com
Certificate SAN https://github.com/kanaworksai-hue/audio-studio-mcp/.github/workflows/release.yml@refs/tags/v1.2.0
Rekor log index 3923098506
Predicate type PyPI publish attestation https://docs.pypi.org/attestations/publish/v1
Subject digest sha256:626d0caed1eadaf675c434ad52640e54fe7da76554af6d21dfcb7cccc4e3745b

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 →

Vulnerabilities 17 findings
ID CVE Severity Vector Fix available
GHSA-g6cj-pr64-35w5 CVE-2026-69247 high yes
GHSA-jwv3-5hgf-82ww CVE-2026-69249 high yes
GHSA-m2h6-j472-rp4c CVE-2026-69248 medium yes
PYSEC-2026-3552 CVE-2026-69247 none yes
PYSEC-2026-3553 CVE-2026-69249 none yes
PYSEC-2026-3554 CVE-2026-69248 none yes
GHSA-63cw-57p8-fm3p CVE-2026-24747 high CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H yes
GHSA-887c-mr87-cxwp CVE-2025-3730 low CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L yes
GHSA-qfhq-4f3w-5fph CVE-2025-3001 medium CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L yes
GHSA-rrmf-rvhw-rf47 CVE-2025-3000 medium CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L yes
GHSA-vgrw-7cvw-pwgx CVE-2025-2999 medium CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L yes
PYSEC-2025-203 CVE-2025-55551 high CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H yes
PYSEC-2025-204 CVE-2025-55552 high CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H yes
PYSEC-2025-206 CVE-2025-55554 medium CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L yes
PYSEC-2026-139 CVE-2026-4538 high CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H no
PYSEC-2026-1970 CVE-2025-3730 low CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:L yes
PYSEC-2026-2286 CVE-2026-24747 high CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H yes

Background: What a vulnerability scan can and cannot prove →

Dependencies 87 packages
Packages resolved 87
Stale 4
No linked repository 1
Tree resolution Complete

Background: SBOMs and build attestations, explained →

MCP tools · 4 exposed · ~1,421 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
edit_audio ~462

Edit existing audio with Stable Audio 3: variation (audio-to-audio), inpainting a time range, or continuation/extension. Returns one WAV path per candidate. Official tip: surrounding context outweighs the prompt — inpaint a large region first and shrink it, and keep prompts plausible for the context.

NameTypeReqDescription
audio_typestring–Same prefixes as generate_audio.
bpm–––
cfgnumber––
duration_seconds––Output length. variation/inpaint default = input length; continue = required total length (> input length).
end_seconds––inpaint only: region end.
input_audiostringyesAbsolute path to an existing audio file (wav/mp3/flac/ogg).
modestringyesvariation = audio-to-audio remix/style or timbre transfer of the whole clip; inpaint = regenerate only [start,end] and keep the rest; continue = extend the clip to duration_seconds, regenerating from…
negative_prompt––Only used when cfg != 1.
original_ideastring–The user's original request, verbatim and in their language. Stored in metadata only.
overlap_secondsnumber–continue only: start regenerating this many seconds before the end of the audible content, so a trailing fade or silence is replaced and the join is smooth.
peak_ceiling_db––Safety ceiling in dBFS: a louder peak is turned down to this, quieter audio is untouched. null disables.
promptstring–What the result (or the regenerated region) should sound like. May be empty for continue.
seed––Base seed for reproducibility; omit or -1 for random.
start_seconds––inpaint only: region start.
stepsinteger––
strengthnumber–variation only: init noise level. ~0.4 keeps melody/rhythm (timbre transfer), ~0.6 style transfer, 0.8+ loose.
title––Short name; used as the file name and stored in metadata. Auto-named if omitted.
variationsinteger–Number of candidates (different seeds).

No output schema declared.

No examples provided.

generate_audio ~429

Generate BGM, instrument loops/stems or sound effects from text with Stable Audio 3 (runs locally). Returns one WAV path per candidate.

NameTypeReqDescription
audio_typestring–music = BGM / full instrumental track; instrument = solo instrument, stem or loop; sfx = sound effect, ambience or one-shot; raw = no prefix (prompt used verbatim)
bpm––Tempo for music/instrument; appended as 'BPM: N.'
cfgnumber–Guidance. 1.0 = official default; 3–7 = stricter prompt adherence (≈2× slower).
duration_seconds––Length in seconds. Default: music 60, instrument 12, sfx 5, raw 30.
negative_prompt––Qualities to avoid. Only takes effect when cfg != 1.
original_ideastring–The user's original request, verbatim and in their language. Stored in metadata only.
peak_ceiling_db––Safety ceiling in dBFS: a louder peak is turned down to this, quieter audio is untouched. null disables.
promptstring–English description (see server instructions). Optional for sfx when the sound_* fields are given.
seed––Base seed for reproducibility; omit or -1 for random.
sound_actionstring–SFX: how it is triggered and evolves over time (attack, decay, movement).
sound_materialstring–SFX: material, texture, timbre.
sound_sourcestring–SFX: what makes the sound (object, instrument, creature, machine).
sound_spacestring–SFX: space and recording (indoor/outdoor, distance, dry/reverberant, mic).
stepsinteger–Sampling steps; 8 is the official default.
title––Short name; used as the file name and stored in metadata. Auto-named if omitted.
variationsinteger–Number of candidates (different seeds) to generate in one call.

No output schema declared.

No examples provided.

process_audio ~499

Post-process and export a clip: trim, trim silence, mono, normalize (peak or LUFS), seamless loop, fades, then write 48 kHz/44.1 kHz WAV, OGG Vorbis or FLAC (+ .json sidecar). The input file is not modified. Steps run in that order. Does not load any model.

NameTypeReqDescription
end_seconds––Trim: keep until here (default: end).
fade_in_msnumber––
fade_out_msnumber––
formatstring–wav = 24-bit PCM; ogg = Ogg Vorbis (Unity/Godot); flac = 24-bit lossless.
input_audiostringyesAbsolute path to an audio file (any format soundfile/ffmpeg reads).
loopboolean–Make a seamless loop by crossfading the tail into the head (output gets shorter by loop_crossfade_ms).
loop_crossfade_msnumber–Loop crossfade: ~50–200 ms for rhythmic BGM cut on a bar, 500–2000 ms for ambience.
monoboolean–Downmix to mono (common for positional game SFX).
normalizestring–none = only the safety ceiling; peak = scale so the peak hits peak_ceiling_db; loudness = EBU R128 to target_lufs, two-pass, linear gain when the true-peak limit allows it, otherwise ffmpeg's dynamic…
ogg_qualitynumber–Vorbis quality 0.1–1.0 (0.6 ≈ 190 kbps stereo).
peak_ceiling_dbnumber–Peak ceiling / target in dBFS.
sample_rateinteger–48000 for video; 44100 keeps the model's native rate (no resampling).
silence_threshold_dbnumber––
start_secondsnumber–Trim: keep from here.
target_lufsnumber–loudness mode: -14 streaming/video, -16 game BGM, -23 broadcast. SFX are usually peak-normalized instead.
title––Output file name stem. Default: <input>_proc.
trim_silenceboolean–Cut leading/trailing silence (useful for one-shot SFX).

No output schema declared.

No examples provided.

unload_models ~31

Free the ~5–7 GB of memory held by the loaded models now (they reload on the next call).

Input schema present but exposes no named parameters.

No output schema declared.

No examples provided.

Install

How do I install the Audio Studio MCP server?

Audio Studio runs locally as a PyPI package, launched with uvx --python 3.11 audio-studio-mcp. The registry lists runner options for this package that are not in this command (they are not ones we reproduce, or they name another package). Check the registry entry before relying on it. Ready-made configuration for Claude, Cursor, VS Code, Codex and 3 more is on this page, copied from each client's own documentation.

pypi · audio-studio-mcp

# add to Claude Code
claude mcp add kanaworksai-hue-audio-studio-mcp -- uvx --python 3.11 audio-studio-mcp

The registry lists runner options for this package that are not in this command (they are not ones we reproduce, or they name another package). Check the registry entry before relying on it.

Common questions

What is the Audio Studio MCP server?

Audio Studio is an MCP server listed in the public MCP registry as io.github.kanaworksai-hue/audio-studio-mcp. Local AI BGM and sound effects with Stable Audio 3 (Apple Silicon MLX, NVIDIA CUDA). This page covers its PyPI package (audio-studio-mcp).

Is the Audio Studio MCP server safe to use?

Audio Studio scores 71 out of 100 on VerifyMCP. We recorded 17 known advisories against it as of 22 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 Audio Studio MCP server expose?

Audio Studio exposes 4 tools: generate_audio, edit_audio, process_audio, unload_models. Their descriptions and schemas cost roughly 1,421 tokens of context every time the server is loaded.

Is the Audio Studio MCP server still maintained?

Audio Studio is still listed as active in the MCP registry. We last reached this channel on 22 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 Audio Studio MCP server under?

Audio Studio declares the MIT 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 Audio Studio MCP server?

VerifyMCP ran 15 checks on Audio Studio, covering Malware scan, Known CVEs, Install scripts, Dependency health and 11 more. One item could not be completed, a gap on our side rather than a finding about Audio Studio: Stability. One category scores 0 for that reason, because we only credit what we can confirm. 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.