User Scanner
User Scanner 未来与实践项目：构建隐私保护的 OSINT 证据实验室
以版本化模块、同意收据、透视图、置信标签、脱敏报告、离线夹具和 MCP 治理为主线。
你将学会
- Explain module, pivot, report, and MCP layers
- Run a bounded scan and interpret uncertainty
- Apply authorization, rate-limit, and data-retention controls
开始前需要
- Basic Git and command-line usage
- Comfort reading a project README
You can explain the project, run its documented first step, and decide what to verify before adopting it.
先看结论
- Consent and evidence receipts can make broad OSINT workflows reviewable.
- Offline fixtures, status semantics, pivot-cycle tests, and redaction enable safer evolution.
- Accessible text/SVG remains canonical when interactive graph views are added.
Make evidence the product
为案例保存授权、模块快照、代理策略、状态、pivot 边、报告哈希、复核人和留存；用离线夹具验证状态、环路、脱敏和导出一致性，文本/SVG 为规范，Three.js 仅作可选图谱。
New vectors, breach integrations, and agent features should remain proposals until their consent, data handling, parser tests, and failure modes are documented.
Define the case manifest
Store package/commit, target type and normalized hash, authorization reference, purpose, module snapshot, proxy/network policy, depth/concurrency, response status, evidence URLs/identifiers, pivot edges, report hashes, reviewer, retention, and deletion receipt. Keep raw personal data and credentials outside the manifest behind access controls.
Generate accessible Markdown/JSON plus local SVG graphs; an optional Three.js explorer can help navigate large pivot graphs but must not replace text evidence or expose hidden data by default.
Ship bounded milestones
Milestone one adds offline module fixtures and distinct found/not-found/blocked/timeout states. Milestone two validates cross-scan depth, cycle prevention, redaction, and report parity. Milestone three adds MCP capability manifests, approvals, and spend/rate budgets. Milestone four canaries new modules and regional changes with human review.
Start report-only, then advisory, then enforce consent, evidence, and retention gates after operators can reproduce them.
Success criteria
A reviewer should replay a safe case, understand every pivot, verify representative results, produce a minimal report, and delete the data on schedule. That is more durable than simply adding more platform modules.
如何选择
| 比较维度 | 方案 A | 方案 B |
|---|---|---|
| Best when | You need predictable behavior and easy auditing | You need adaptive optimization and have reliable telemetry |
| Main risk | May leave performance on the table | Can become difficult to explain or debug |
实施步骤
- 1
Define consent/case/module/pivot/report manifest fields.
- 2
Add offline fixtures for statuses, cycles, redaction, and export parity.
- 3
Govern MCP capabilities, approvals, budgets, and regional changes.
- 4
Move from report-only to advisory/enforced evidence and retention gates.
可复制示例
case:
package_commit: <sha>
target_sha256: <hash>
authorization_ref: self-audit-001
modules: [github, etsy]
cross_depth: 1
evidence: [results.json]
report_sha256: <hash>
retention_until: 2026-10-01常见问题
Should the future lab store full profile data in Git?
No. Keep controlled raw evidence outside Git and commit only sanitized fixtures, hashes, and manifests.
Is a 3D pivot graph required?
No. Use accessible text and local SVG first; add Three.js only as an optional exploration layer.
资料来源
- User Scanner README (captured 2026-08-31)来源核查 2026-08-31
- User Scanner repository来源核查 2026-08-31