Prompt catalog · v0.1 preview
The prompts your teams are running, inventoried and kept current.
Scriven catalogs every Claude Project, custom GPT, Gemini Gem, and standalone prompt in use across your non-engineer teams. Each entry is cross-checked against your authoritative sources, so drift shows the moment it appears.
The problem
A library has accumulated. Nobody has catalogued it.
01 / COUNT
Artifacts multiply
A five-hundred-person company already runs hundreds of Projects, GPTs, Gems, and prompts. Nobody has counted them. Nobody has named most of them twice.
02 / STEWARDSHIP
Nobody owns the shelf
Individual users create. Individual users leave. The artifacts remain behind, used daily by teammates who inherited them and would not think to ask where they came from.
03 / FIDELITY
Sources shift. References do not.
The brand guide updates. The pricing sheet is revised. Forty marketing prompts continue to reference last year's version, and nobody knows to look.
What Scriven catalogs
Five artifact classes. Matched against the sources you name.
- Claude Project
- instructions · knowledge files · system prompt
- your designated sources
- Custom GPT
- instructions · linked files · retrieval behavior
- your designated sources
- Gemini Gem
- system prompt · connected resources
- your designated sources
- Standalone prompt
- shared prompt text · referenced snippets
- your designated sources
- Workspace totals
- drift rate · unowned percentage · source fidelity
- last catalog snapshot
Catalog
The cataloger's table, viewed from above.
One card per artifact. Oxblood marks an entry that has drifted from its source.
catalog / marketing · 312 artifacts
3 drifted · 2 current
How it works
Four steps from connection to a catalog kept current.
- 01
Connect
Read-only access to Claude, ChatGPT, Gemini, and the shared prompt workspaces your teams already use. - 02
Catalog
Every artifact entered with its type, owner, last change, and the sources it references. Nothing is created or edited. - 03
Cross-check
Each artifact is matched back to the source documents you designate as authoritative. - 04
Report
Drift is flagged. The catalog stays in fidelity to the library beneath it, week after week.
Trust
What we read. What we retain. What we will not do.
- Read-only, always
- We do not write, edit, or republish any artifact. The catalog is descriptive, not prescriptive.
- Contents analyzed, not stored
- Prompt text is analyzed to detect references and drift, then discarded. Long-term storage is opt-in per workspace.
- Your region, your data
- Findings return to you. Nothing is retained on our side beyond catalog metadata.
- SOC 2 path in progress
- Type I target: Q4 2026. Type II follows. We will not claim certification we do not hold.
Pricing
One tier while we build. Enterprise when you need it.
Starter
- Up to 250 catalogued artifacts
- Monthly cross-check, delivered as report
- Three designated source documents
- Email support, direct line to the team
Enterprise
- Unlimited artifacts and sources
- Continuous cross-check with drift alerts
- SSO, custom retention, regional residency
- Shared channel with our team
Run inventory
Open the catalog.
Early access is free while we build v0.1. You receive the first catalog and a direct line to the team.