Trusted — Risk Score 0/100
Last scan:2 days ago Rescan
0 /100
felo-superAgent
AI conversation with real-time SSE streaming on a persistent LiveDoc canvas
Felo SuperAgent is a legitimate API client skill that makes authenticated HTTP requests to the Felo Open Platform API. All capabilities are properly declared, no sensitive data access, no credential exfiltration, and no suspicious execution patterns detected.
Skill Namefelo-superAgent
Duration21.1s
Enginepi
Safe to install
No action needed. The skill is safe to use as described.
ResourceDeclaredInferredStatusEvidence
Network READ READ ✓ Aligned HTTP requests to https://openapi.felo.ai with Bearer token authentication (run_s…
Shell NONE NONE SKILL.md: Shell execution not declared. Scripts use Node.js for API calls only, …
Filesystem NONE NONE No file read/write operations in scripts
Environment READ READ ✓ Aligned SKILL.md declares env var usage. Scripts read FELO_API_KEY only (run_superagent.…
Clipboard NONE NONE No clipboard access detected
Browser NONE NONE No browser access detected
Database NONE NONE No database access detected
Skill Invoke READ READ ✓ Aligned Depends on felo-livedoc skill for LiveDoc management (clawhub.json)
1 High 7 findings
🔑
High API Key 疑似硬编码凭证
API_KEY="your-api-key-here"
README.md:64
🔗
Medium External URL 外部 URL
https://felo.ai/livedoc/...
README.md:266
🔗
Medium External URL 外部 URL
https://custom-api.example.com
README.md:376
🔗
Medium External URL 外部 URL
https://openapi.felo.ai/docs/api-reference/v2/superagent.html
README.md:394
🔗
Medium External URL 外部 URL
https://openapi.felo.ai/docs/
README.md:395
🔗
Medium External URL 外部 URL
https://felo.ai/livedoc/QPetunwpGnkKuZHStP7gwt
SKILL.md:375
🔗
Medium External URL 外部 URL
https://openapi.felo.ai
SKILL.md:644

File Tree

5 files · 77.6 KB · 2005 lines
Markdown 2f · 1208L JavaScript 2f · 784L JSON 1f · 13L
├─ 📁 scripts
│ ├─ 📜 run_style_library.mjs JavaScript 213L · 6.5 KB
│ └─ 📜 run_superagent.mjs JavaScript 571L · 20.4 KB
├─ 📋 clawhub.json JSON 13L · 793 B
├─ 📝 README.md Markdown 398L · 12.7 KB
└─ 📝 SKILL.md Markdown 810L · 37.2 KB

Security Positives

✓ No shell command execution - uses Node.js for all API interactions
✓ No credential exfiltration - FELO_API_KEY used only for Bearer authentication to legitimate API endpoint
✓ No sensitive path access - does not read ~/.ssh, ~/.aws, or .env files
✓ No base64/encoded payloads piped to shell
✓ No remote script execution (curl|bash, wget|sh)
✓ No hidden instructions in HTML comments or embedded data
✓ API key read from environment variable only, never hardcoded in code
✓ Clear doc-to-code alignment: SKILL.md accurately describes all functionality
✓ Scripts are self-contained JavaScript with no external dependencies requiring installation
✓ SSE streaming is standard HTTP behavior for real-time responses
✓ All external URLs point to legitimate Felo AI domain (openapi.felo.ai)