Willder
Willder is a shared, access-controlled memory for AI agents that connects over MCP so several agents and people can shar
Link is nofollow until the page is claimed
Is this your product?
Claim Willder and this page fills itself.
Right now it shows a scraped screenshot and a few lines. Claimed pages get a permanent dofollow link, a screenshot gallery, features, pricing, FAQ, your founder story, and the questions visitors ask.
Not yours, or want it gone? Remove this page.
Homepage screenshot taken by Automateo. Claimed pages show the founder's own screenshots. Claim and replace it
About Willder
From the product's websiteAgents connect to Willder's MCP server with their own scoped, expiring token and read/write facts, decisions and rules. Every read and write passes one database permission check logged to an append-only audit log; revoked tokens are refused on the next call. Facts are versioned, stale writes are refused, and new sessions get a task-start brief from their scope. Agents start from zero, redo each other's work, overwrite each other's facts, and can see data they shouldn't. Built for: Individuals and teams running multiple AI agents (e.g. Claude Code, Codex) who need shared context with access control. Main features: Scope tree with per-agent tokens (read/write/admin), 7-day default, 90-day max; Permissions enforced by database query, not prompt; Append-only audit log of every allow/deny decision; Fact versioning with stale_version write refusal; Task-start brief and change notices; Claude Code session hook.
This is what Willder's page looks like claimed
Everything below fills in from the founder's answers or their ListingBro Listing Kit. Nothing to write twice.
- Features, use cases and FAQ, marked up for search
- Pricing, free trial and launch offer
- Founder bio, maker comment and interview story
- Visitors' questions sent to you, with their email
- A permanent dofollow link from a DR 35 site
