Open-source alternative to Claude Cowork — a local-first AI agent desktop app · multi-model · self-evolving skills · privacy-first
C elevated
- capability exposure inferred + 35
- recent drift inferred + 20
- tool safety inferred + 5
- 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 6m ago · see ecosystem CVEs →
- C · 49 → C · 57
Inferred surface — each links to servers holding it:
No known CVEs for this server.
- medium dangerous code
dynamic exec: eval(), new Function()
prompt-surface
shipped agent-instruction files + hidden-content / dangerous-code findings —
quoted from the analyzed source
analyzed commit f4b861f · analyzer v28 · 1d ago
skills & prompt files 29
- agent-rules PM-Shawn-Abu-Cowork-f4b861f/CLAUDE.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/Abu-Browser/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/alert-sop/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/algorithmic-art/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/brand-guidelines/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/canvas-design/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/claude-api/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/create-agent/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/doc-coauthoring/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/docx/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/frontend-design/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/html-widget/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/infographic/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/init/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/internal-comms/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/mcp-builder/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/mermaid-diagram/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/pdf/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/pptx/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/reflect/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/schedule/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/skill-creator/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/slack-gif-creator/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/svg-diagram/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/theme-factory/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/trigger/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/web-artifacts-builder/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/webapp-testing/SKILL.md
- skill PM-Shawn-Abu-Cowork-f4b861f/builtin-skills/xlsx/SKILL.md
danger signals5
- dynamic code execution eval() PM-Shawn-Abu-Cowork-f4b861f/abu-chrome-extension/src/background/index.ts :489
return eval(jsCode); - dynamic code execution eval() PM-Shawn-Abu-Cowork-f4b861f/src-tauri/browser-extension/background.js :370
return eval(jsCode); - dynamic code execution new Function() PM-Shawn-Abu-Cowork-f4b861f/src/components/chat/widgetReceiverDom.test.ts :57
const buildApi = new Function(`${WIDGET_RECEIVER_DOM_JS} - suspicious endpoint 169.254.169.254 (cloud metadata) PM-Shawn-Abu-Cowork-f4b861f/src/core/tools/definitions/webTools.test.ts :28
const result = await httpFetchTool.execute({ url: 'http://169.254.169.254/latest/meta-data/' }); - suspicious endpoint 100.100.100.200 (cloud metadata) PM-Shawn-Abu-Cowork-f4b861f/src/core/tools/definitions/webTools.test.ts :38
const result = await httpFetchTool.execute({ url: 'http://100.100.100.200/latest/meta-data/' });
- recent drift +20 capability drift →
[](https://mcpobservatory.com/servers/github:PM-Shawn/Abu-Cowork/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 PM-Shawn.