How we rank
Every tool is researched & quality-gated
Cited·Index is a curated index, not a scraped one. Before any tool appears here it must pass an automated quality gate. The same checks are enforced in code, so a thin or unverified page is structurally impossible to publish. Here is exactly how a listing earns its place.
The quality gate
Every listing must clear four bars before it can be published:
- Structured research. A minimum set of structured fields (what it is, what it does, key features, pricing and real use cases) must be present and complete. A stub can't pass.
- Sourced facts. Claims are backed by cited sources, not marketing copy. Pricing and capabilities reflect what the tool actually does.
- On-topic fit-check. A tool that isn't genuinely relevant to this directory is rejected, however complete its record. Relevance is a gate, not a nice-to-have.
- Freshness. Listings carry a last-verified date and are re-checked over time; the index you're reading was last updated August 18, 2026.
The editorial score
Every published tool wears a single 0-10 editorial score. It isderived, never asserted: each part is computed from verifiable structure already on the page (how the facts are sourced, which weaknesses are admitted, how deep and how legible the record is), so the number can be defended line by line, and it cannot be bought(paid placement feeds nothing into it). These are the exact weights the engine uses:
| Dimension | Weight | What it measures |
|---|---|---|
| Evidence | 25% | How many independent domains the facts on this page cite, and how corroborated they are. |
| Balance | 19% | How even-handed the verdict is beyond the bare minimum: extra genuine cons, real limitations, and a specific "who it is NOT for". |
| Depth | 18% | How much substantive structure is filled — features, use cases, FAQ, integrations and how-it-works, counted as a whole. |
| Pricing transparency | 15% | Whether pricing is legible: named tiers, a concrete price anchor, a disclosed free trial. |
| Maturity | 13% | Signals it is a real, buildable product: a public API, first-party docs, a source repo, and a documented method. |
| Freshness | 10% | How recently a verification pass re-checked the facts (decays as the record ages). |
The overall score maps to a grade:
- Exceptional 8.5-10.0
- Strong 7.0-8.4
- Solid 5.0-6.9
- Fair 4.0-4.9
- Thin below 4.0
The same score is shown on the tool's own page and on every grid card (computed one way, from one source of truth) and it moves only when re-research changes the underlying facts.
How listings are ordered
Within a category, listings are ordered by editorial tier and then alphabetically, never by who paid. Any commercial placement is labelled as such. The label looks like this: Sponsored, and still has to pass the very same quality gate; money can move where a tool appears, never whether it earned a page.
Corrections & submissions
Anyone can submit a tool for review, but a submission is only ever a candidate. It enters the same research-and-gate pipeline before it can be published, and nothing a submitter writes becomes page copy unverified. Spotted something wrong? Tell us and we'll re-verify.