Scan Report
5 /100
agent-profile-images
Agent Profile Images for OpenClaw Control UI — upload custom avatars, generate themed AI profile images
This is a legitimate UI feature package for agent profile images containing only documentation and TypeScript reference implementations. The flagged base64 decode is standard API response handling (OpenAI b64_json), all functionality is documented, and no malicious behavior is present.
Safe to install
No action required. The skill is safe for use.
| Resource | Declared | Inferred | Status | Evidence |
|---|---|---|---|---|
| Filesystem | WRITE | WRITE | ✓ Aligned | Avatar files written to agent workspace avatars/ directory |
| Network | READ | READ | ✓ Aligned | Calls OpenAI API https://api.openai.com/v1/images/generations |
| Shell | NONE | NONE | — | No shell execution found |
| Environment | NONE | READ | ✓ Aligned | Reads API keys from env (ANTHROPIC_API_KEY, OPENAI_API_KEY) - necessary for feat… |
1 Critical 2 findings
Critical Encoded Execution Base64 编码执行（代码混淆）
Buffer.from(b64, "base64" references/src-gateway-server-methods-agents-ts.txt:534 Medium External URL 外部 URL
https://docs.openclaw.ai references/ui-src-ui-app-render-ts.txt:603 File Tree
17 files · 409.8 KB · 11567 lines Text 16f · 11446L
Markdown 1f · 121L
├─
▾
references
│ ├─
src-gateway-method-scopes-ts.txt
Text
│ ├─
src-gateway-protocol-index-ts.txt
Text
│ ├─
src-gateway-protocol-schema-agent-ts.txt
Text
│ ├─
src-gateway-protocol-schema-agents-models-skills-ts.txt
Text
│ ├─
src-gateway-protocol-schema-protocol-schemas-ts.txt
Text
│ ├─
src-gateway-protocol-schema-types-ts.txt
Text
│ ├─
src-gateway-server-methods-agent-ts.txt
Text
│ ├─
src-gateway-server-methods-agents-ts.txt
Text
│ ├─
src-gateway-server-methods-list-ts.txt
Text
│ ├─
src-gateway-session-utils-ts.txt
Text
│ ├─
ui-src-ui-app-render-ts.txt
Text
│ ├─
ui-src-ui-app-ts.txt
Text
│ ├─
ui-src-ui-app-view-state-ts.txt
Text
│ ├─
ui-src-ui-types-ts.txt
Text
│ ├─
ui-src-ui-views-agents-panels-overview-ts.txt
Text
│ └─
ui-src-ui-views-agents-ts.txt
Text
└─
SKILL.md
Markdown
Security Positives
✓ All functionality clearly documented in SKILL.md
✓ Proper path guards prevent directory traversal attacks
✓ File writes restricted to agent workspace directory
✓ Image size validation (2MB limit) prevents DoS
✓ API key access is scoped to OpenAI/Anthropic for avatar generation only
✓ No shell execution or subprocess calls
✓ No credential exfiltration - keys used only for intended API calls
✓ Base64 usage is standard API pattern for image data transfer
✓ Proper input validation and sanitization throughout