Local-first AI agent workspace for coding, writing, design, research, and automation — one runtime for desktop GUI and TUI.
- capability exposure inferred + 35
- recent drift inferred + 12
- tool safety inferred + 12
- trust mitigators mixed − 3
inferred mixed
The A–E grade is our heuristic synthesis — a "review this" prompt, not a verdict. Each factor is tagged by what backs it: attested (a verifiable record), reported (a third party's claim), or inferred (our own heuristic, e.g. permissions). See methodology.
graded 1m ago · see ecosystem CVEs →
No known CVEs for this server.
- high dangerous code
committed secret: Google OAuth client secret · dynamic exec: new Function()
analyzed commit 1a6c8f6 · analyzer v28 · 2h ago
skills & prompt files 14
- skill KunAgent-Kun-1a6c8f6/.claude/skills/openspec-apply-change/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.claude/skills/openspec-archive-change/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.claude/skills/openspec-explore/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.claude/skills/openspec-propose/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.codex/skills/openspec-apply-change/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.codex/skills/openspec-archive-change/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.codex/skills/openspec-explore/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.codex/skills/openspec-propose/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.cursor/skills/openspec-apply-change/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.cursor/skills/openspec-archive-change/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.cursor/skills/openspec-explore/SKILL.md
- skill KunAgent-Kun-1a6c8f6/.cursor/skills/openspec-propose/SKILL.md
- agent-rules KunAgent-Kun-1a6c8f6/AGENTS.md
- agent-rules KunAgent-Kun-1a6c8f6/docs/AGENTS.md
danger signals9
- dynamic code execution new Function() KunAgent-Kun-1a6c8f6/src/main/services/ui-plugin-cdp-theme-controller.test.ts :88
const runFixedRuntime = new Function('document', `return ${parameters.expression}`) - dynamic code execution new Function() KunAgent-Kun-1a6c8f6/src/renderer/src/components/design/canvas/HtmlFrameOverlay.test.ts :115
const execute = new Function( - dynamic code execution new Function() KunAgent-Kun-1a6c8f6/src/renderer/src/design/design-html-quality.runtime-design-quality-findings.test.ts :204
expect(new Function(`return ${script}`)()).toEqual([]) - dynamic code execution new Function() KunAgent-Kun-1a6c8f6/src/renderer/src/design/design-token-extract.test.ts :51
const execute = new Function(`return ${TOKEN_EXTRACT_GUEST_SRC}`) - suspicious endpoint 169.254.169.254 (cloud metadata) KunAgent-Kun-1a6c8f6/kun/src/services/extension-account-broker.test.ts :297
deviceAuthorizationUrl: 'https://169.254.169.254/device' - suspicious endpoint 169.254.169.254 (cloud metadata) KunAgent-Kun-1a6c8f6/kun/tests/web-tool-provider.test.ts :423
'http://169.254.169.254/latest/meta-data/', - suspicious endpoint api.telegram.org KunAgent-Kun-1a6c8f6/src/main/telegram-runtime.ts :24
const TELEGRAM_API_BASE = 'https://api.telegram.org' - committed secret Google OAuth client secret KunAgent-Kun-1a6c8f6/kun/src/adapters/model/gemini-cli-oauth.ts :15
GOCSPX…(35 chars, redacted) - committed secret Google OAuth client secret KunAgent-Kun-1a6c8f6/kun/src/adapters/model/gemini-code-assist-model-client.ts :24
GOCSPX…(35 chars, redacted)
- recent drift +12 capability drift →
[](https://mcpobservatory.com/servers/github:KunAgent/Kun/security) Heuristic, inferred signals — false positives (legitimately powerful tools, forks, language ports) are expected. Treat each as "review this", not a verdict. See the ecosystem-wide picture on the security hub, or the fleet security of KunAgent.