Waxell

skill

stepfun-asr High risk

daymadedaymade/claude-code-skillsupdated

Transcribes Chinese/English audio with StepFun's stepaudio-3-asr-max via its SSE endpoint (not /v1/audio/transcriptions), one call handles long-form audio with no chunking. Use when migrating from step-asr/stepaudio-2.5-asr, or hitting the misleading "model not supported" error (actually wrong endpoint). Triggers on 阶跃 ASR, 语音识别. Not for TTS with the sibling model (use stepfun-tts).
daymade's own description, from the skill's files.
Stars
1,534
Version
v2
Copies
0

Overview

blockHigh risk

Waxell rates it high risk: 16 findings, highest severity critical.

Highest severity
critical
Findings
16
What ran
Static analysis + deep scan

A clean result means nothing alarming in what is published, not a guarantee. How the checks work

We don't republish the skill's instructions. Read them on GitHub at the reviewed commit.

Security

Security scan

Checked yesterday

Result

High risk Waxell rates it high risk: 16 findings, highest severity critical.

Checked with static analysis + deep scan: servers it launches, packages it installs at runtime, scripts piped to a shell and bundled binaries. The deep scan also read the instructions for prompt injection and data exfiltration. Nothing was run. How the checks work

Findings (16)

critical · 2

  • critical

    Data Flow: Tainted flow: 'req' from os.environ.get (line 42, credential/environment) → urllib.request.urlopen (network o

    scripts/asr_file.py:46 · skillspector:TT3

  • critical

    Data Flow: Tainted flow: 'req' from os.environ.get (line 162, credential/environment) → urllib.request.urlopen (network

    scripts/asr_transcribe.py:174 · skillspector:TT3

high · 3

  • high

    Anti-Refusal: Anti-Refusal Statement

    references/known_issues.md:106 · skillspector:AR2

  • high

    Dangerous Code Execution: os.system() or os exec-family call

    scripts/external_dispatch.py:203 · skillspector:AST5

  • high

    Data Exfiltration: Env Variable Harvesting

    scripts/external_dispatch.py:203 · skillspector:E2

medium · 11

  • medium

    Without declared permissions the skill's intent is opaque and cannot be validated.

    SKILL.md:1 · skillspector:LP3

  • medium

    Data Exfiltration: External Transmission

    SKILL.md:21 · skillspector:E1

  • medium

    Data Exfiltration: External Transmission

    SKILL.md:42 · skillspector:E1

  • medium

    Data Exfiltration: External Transmission

    references/api_reference.md:8 · skillspector:E1

  • medium

    Data Exfiltration: External Transmission

    references/api_reference.md:102 · skillspector:E1

  • medium

    Data Exfiltration: External Transmission

    references/known_issues.md:70 · skillspector:E1

  • medium

    Data Exfiltration: External Transmission

    references/known_issues.md:180 · skillspector:E1

  • medium

    Data Exfiltration: External Transmission

    scripts/asr_file.py:25 · skillspector:E1

  • medium

    Data Exfiltration: External Transmission

    scripts/asr_transcribe.py:5 · skillspector:E1

  • medium

    Data Exfiltration: External Transmission

    scripts/asr_transcribe.py:40 · skillspector:E1

  • medium

    Data Exfiltration: External Transmission

    tests/test_routes.py:18 · skillspector:E1

Versions

Version history

Each distinct content we have seen, newest first, with the verdict it got at the time.

  1. v22026-11-161 added, 2 changed High riskgot riskier
  2. v12026-11-03 No issues

Files

Files

8 files, 72 KB in total. Names only; open them on GitHub.

  • SKILL.md
  • references/api_reference.md
  • references/known_issues.md
  • scripts/asr_file.py
  • scripts/asr_transcribe.py
  • scripts/check_params.py
  • scripts/external_dispatch.py
  • tests/test_routes.py

See and control which skills your people use

Inside your organization, Waxell scans each skill and plugin before anyone installs it, approves what passes and publishes it to a private marketplace, and puts every MCP tool call behind one governed URL.

30 minutes. No commitment.

SOC 2 Type II attested · ISO/IEC 27001:2022 certified · HIPAA with a BAA · aligned to the EU AI Act and NIST AI RMF

More images, video & audio skills and plugins

See all