中文 EN

Scan Report

Scan New Files

✓ Trusted — Risk Score 5/100
Last scan:16 hr ago Rescan
5 /100
meeting-minutes-assistant
会议录音转结构化纪要 (ASR → LLM总结 → 可选TTS)
The skill is a straightforward meeting minutes assistant with no hidden or malicious behavior—code performs exactly what documentation describes.
Skill Namemeeting-minutes-assistant
Duration34.2s
Enginepi
✓
Safe to install
No action needed. The skill can be safely deployed.

Findings 1 items

Severity Finding Location
Low
Dependencies not pinned to exact versions Supply Chain
requirements.txt specifies openai>=1.30.0, requests>=2.31.0, python-dotenv>=1.0.0. While these are known legitimate packages, loose versioning allows a future compatible-but-malicious version to be installed on pip install.
openai>=1.30.0
requests>=2.31.0
python-dotenv>=1.0.0
→ Pin to exact versions (e.g., openai==1.30.0) to ensure reproducible builds.
 requirements.txt:1
ResourceDeclaredInferredStatusEvidence
Filesystem READ READ ✓ Aligned audio_file.open('rb') at scripts/run_meeting_minutes.py:75
Filesystem WRITE WRITE ✓ Aligned OUTPUT_DIR.mkdir() and output_path.write_text() at lines 196, 218
Network READ READ ✓ Aligned requests.post() to ASR/TTS URLs at lines 63, 69
Shell NONE NONE — No subprocess or os.system calls found
Environment NONE READ ✓ Aligned Reads .env via python-dotenv; not declared in SKILL.md but benign
4 findings
🔗
Medium External URL 外部 URL
https://models.audiozen.cn/v1
SKILL.md:112
🔗
Medium External URL 外部 URL
https://api.senseaudio.cn
SKILL.md:113
🔗
Medium External URL 外部 URL
https://api.senseaudio.cn/v1/audio/transcriptions
scripts/run_meeting_minutes.py:63
🔗
Medium External URL 外部 URL
https://api.senseaudio.cn/v1/t2a_v2
scripts/run_meeting_minutes.py:69

File Tree

6 files · 15.0 KB · 506 lines
Python 1f · 313L Markdown 3f · 188L Text 1f · 3L YAML 1f · 2L
├─ ▾ 📁 agents
│ └─ 📋 openai.yaml YAML 2L · 51 B
├─ ▾ 📁 references
│ ├─ 📝 output_schema.md Markdown 29L · 845 B
│ └─ 📝 prompts.md Markdown 40L · 792 B
├─ ▾ 📁 scripts
│ └─ 🐍 run_meeting_minutes.py Python 313L · 10.2 KB
├─ 📄 requirements.txt Text 3L · 53 B
└─ 📝 SKILL.md Markdown 119L · 3.0 KB

Dependencies 3 items

PackageVersionSourceKnown VulnsNotes
openai >=1.30.0 pip No Version not pinned; known legitimate package
requests >=2.31.0 pip No Version not pinned; known legitimate package
python-dotenv >=1.0.0 pip No Version not pinned; known legitimate package

Security Positives

✓ No subprocess, os.system, or shell execution found
✓ No credential harvesting or environment variable enumeration for theft
✓ No base64, eval, or code obfuscation
✓ All network calls target only declared API endpoints (SenseAudio, LLM backend)
✓ Code-to-doc alignment is high: every function maps to documented behavior
✓ No access to sensitive paths (~/.ssh, ~/.aws, .env secrets exfiltration)
✓ No data exfiltration or C2 communication patterns
✓ File writes scoped to skill-local outputs/ directory only