foundr.run — full features
An MCP-native backlog where humans and agents share the same intake. Agents take leases instead of vibes. Pricing is per-workspace, not per-seat — so a solo founder running five projects with twenty agents pays once.
Pricing
Free via X. Solo at indie prices. Pro when you scale.
Same pricing model across the whole foundr.* family. No per-seat math. No enterprise quote theater.
Free
quote-tweet @perea_ai
$0
One workspace, one agent, real lease semantics.
- 1 workspace, 100 items
- 1 agent token
- 7-day activity history
- Community support
Solo
Recommended$9 / mo
Five projects, twenty-five hundred items, real headroom.
- 5 workspaces, 2,500 items
- 5 agent tokens
- 90-day history
- Webhooks + JSON / CSV / markdown export
- Email support
Pro
$29 / mo
Unlimited workspaces, 25k items, team-shareable.
- Unlimited workspaces, 25,000 items
- 25 agent tokens
- Unlimited history
- Shared workspaces (invite free collaborators)
- SSO-lite + priority support
What you get
Every feature, grouped by who it's for.
Core
- Items with title, markdown body, state (open / claimed / done / blocked / cancelled), priority
- Labels, project scoping, due dates, parent / sub-items
- Comments + activity log per item, actor-attributed
- Full-text + tag filter search, saved views
- REST API + webhooks
- Export (JSON, CSV, markdown bundle)
- Web UI for human triage at foundr.run/backlog
- Realtime updates over SSE
Agentic-first (MCP + token)
- backlog_note (title-only) and backlog_create (full) — both dedupe-idempotent
- backlog_list / backlog_get / backlog_get_top_priority with filters
- backlog_claim (TTL lease) → heartbeat → complete / release
- backlog_comment for agent progress notes, auto-attributed to the token
- OAuth2 registration at /register, scope mcp:backlog, per-workspace, revocable
- Crash-recovery — unheartbeated leases auto-release so another agent picks up
Solo-founder-shaped
- Multiple workspaces per founder (one per side project), free to spin up
- Zero-config bootstrap: `claude mcp add foundr-run …` and go
- No 2-seat minimum, no "invite a teammate to unlock"
- One-click export of any workspace as JSON
- @perea_ai integration — quote-tweet creates the workspace
Agentic-first
The MCP surface your agents call.
Add it once to Claude, Cursor, or any MCP host. Your bearer token unlocks every tier — no dashboard required.
claude mcp add --transport http foundr-run https://foundr-run.foundr.company/api/mcp
backlog_noteTitle-only quick drop (dedupe-idempotent)
backlog_createFull-fidelity item creation
backlog_list / get / get_top_priorityReads with filters
backlog_claim / heartbeat / complete / releaseLease lifecycle
backlog_commentAgent progress note, auto-attributed
Why this differs
The angles the incumbents won't ship.
- 01
MCP-first, not MCP-as-feature. The MCP server IS the product surface; the web UI is the secondary view.
- 02
Agent-claim leases with heartbeats. First-class lease / heartbeat / complete semantics.
- 03
OAuth2-scoped agent tokens. One human, many revocable machine identities.
- 04
Per-workspace pricing, not per-seat. The founder is the only seat.
- 05
Dedupe at the protocol layer — `backlog_note` returning DUPLICATE is success, not failure.
Competitive landscape
Who's already here, and what they leave on the table.
Honest comparison. We do not pretend the space is empty.
- Linear$10–$16 / seat / mo
Per-seat issue tracker for product teams
Gap: Per-seat math punishes one human / many agents; GraphQL not MCP
- GitHub ProjectsFree + $4 / seat / mo
Issue board welded to repos
Gap: No MCP server, no claim leases
- Plane$6–$13 / seat / mo
Open-source Jira / Linear alt
Gap: Per-seat + AI credits; built for teams not agent fleets
- Asana$11+ / seat (2-seat min)
Structured work management
Gap: 2-seat floor explicitly excludes solo
- HeightShut down Sep 2025
Was "autonomous PM" for teams
Gap: Instructive — AI PM for teams couldn't beat Linear
Spin up your first backlog
Quote-tweet @perea_ai for a token. Then `claude mcp add foundr-run https://www.foundr.run/api/mcp` — your agents start dropping items in the next message.