Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsTOS Privacy Audit
Privacy-protective bonus flags cover data minimization, self-hosted analytics, and open-source clients.
$5
TOS Privacy Audit
Example session with this skill installed
Audit Notion's privacy policy and terms. Should I trust them with my notes?
- Read your context and instructions
- Compiled the tos privacy audit
Notion Privacy Policy & ToS Audit
Summary: Notion keeps your notes yours and gives you a full set of privacy rights, but the terms force arbitration and waive class actions, and the policy feeds advertising partners data it admits counts as a sale or share under California law. Final grade: E, 5.75 / 10. Fine for casual notes with an alias email; keep genuinely sensitive material out.
Documents reviewed: Privacy Policy (effective October 1, 2026), Personal Use Terms of Service (effective April 5, 2024; organizations fall under the Master Subscription Agreement). Analysis date: October 14, 2026.
| Category | Weight | Score | What drove it |
|---|---|---|---|
| Data collection & telemetry | 20% | 4/10 | "IP address... MAC address, cookie identifiers... inferred general location," plus cross-device tracking and third-party ad technology, with opt-outs available |
| Data sharing & selling | 25% | 5/10 | "We do not use your information to advertise any third party's products," but advertising partners receive identifiers and browsing data, which Notion concedes "may be considered a 'sale' of personal information under applicable law" |
| User rights & control | 15% | 8/10 | Access, correction, deletion, restriction, opt-out, appeal, authorized agents; Global Privacy Control signals honored |
| Retention & security | 10% | 4/10 | Kept "for as long as you use our Services or as necessary"; de-identified data is exempt from the policy and usable "for any other legally permissible purposes"; standard security with breach notice |
| Legal terms & ToS fairness | 15% | 5/10 | "any Claim will be settled by final and binding arbitration" (JAMS) with waiver of "trial by jury or... class action"; terms changes get about 30 days notice |
| Third parties & subprocessors | 5% | 8/10 | Public subprocessor list referenced; processor categories named in the policy |
| Service terms & general quality | 10% | 7/10 | "We do not claim any ownership rights in any User Content" with a license limited to operating the service; caveat: "some of your User Content may not be completely removed and copies... may continue to exist"; refund terms not addressed in the reviewed text |
Flags: +0.25 GDPR/CCPA protections beyond the minimum market (supervisory-authority complaint right, Data Processing Addendum, transfer impact assessments). No penalty flags with direct evidence in the reviewed text.
Final weighted score: 5.75 / 10, Grade E
Next steps:
- Keep using it for notes with an alias email; the ownership and rights terms are genuinely good
- Turn on Global Privacy Control in your browser or click the "Do Not Sell or Share My Info" link to opt out of the advertising-partner sharing
- Keep passwords, financial details, and sensitive personal material out of your workspaces: copies can persist after deletion and de-identified data escapes the policy
- Re-run the audit before high-stakes decisions; policies change (this one took effect October 1, 2026)
Connects securely to your tools. The creator never sees your data.
About this skill
Paste a ToS or name a service. Your agent fetches the documents, scores them against a fixed 7-category weighted rubric, quotes the exact clause behind every score, and hands you a letter grade plus a verdict you can act on. Sign up freely, use an alias email, or walk away.
What it scores (weighted rubric):
- Data collection & telemetry (20%): what they take, how they watch you
- Data sharing & selling (25%): "share with partners for value" counts as selling
- User rights & control (15%): export, deletion, no dark patterns
- Retention & security (10%): how long they keep it, what they promise
- Legal terms & ToS fairness (15%): forced arbitration, unilateral changes, "continued use = consent"
- Third parties & subprocessors (5%): named or hidden
- Service terms & general quality (10%): content ownership, refund terms, offboarding traps
Penalty flags cover anti-disparagement clauses, change-without-notice terms, and merger voids. Privacy-protective bonus flags cover data minimization, self-hosted analytics, and open-source clients. Both map to the final A–F grade.
Built-in discipline that makes the output trustworthy:
- Silence in a contract favors the drafter: unaddressed categories score low instead of assumed friendly
- Only the document text counts as evidence; marketing claims do not
- Every score carries a quoted clause and the document's stated version and date
- Comparison mode scores rivals identically, returns a ranked table, and ends with a recommendation
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scanned
Verified clean 5 days ago
- Passed all security checks, Safe to install