Local-First · No Telemetry · BYOK Security
Your AI agents get cheaper and smarter over time
Add your own API key (Anthropic, OpenAI, or Google), connect GitHub, and let AI agents write, review, and ship code while you stay in control.
- Multiple AI agents working at the same time
- Project memory that never forgets your codebase
- Connect GitHub, bring your own API key — free to start
Now in soft launch — Download Free
See it in action
This is the real Tendril UI — every project, agent, and dollar rendered as a live knowledge graph. No mockups, no placeholders.
Pricing
Simple, Honest Pricing
Download free. Add your own API key from Anthropic, OpenAI, or Google. You pay your AI provider directly — Tendril is free.
MonthlyAnnualSAVE 17%
Free Tier
$0/month
Free Forever
- Bring Your Own API Key (BYOK)— Anthropic, OpenAI, or Google — you pay your provider directly
- 1 active project
- 2 parallel sub-agents— same-priority tasks run concurrently
- Full pipeline— discovery, planning, execution, commit
- Diff viewer & code review
- Security monitoring— injection detection, scope enforcement, quarantine
- Auto-verify on edits— syntax check after every file change
- MCP server support
- Community support
Pro Tier
$29/month
Billed monthly, cancel anytime
- Everything in Free
- Unlimited projects (up to 25 active)— archive and start new anytime
- Unlimited parallel agents— no Tendril cap — limited only by your hardware + API
- Knowledge Graph— indexes your codebase, cuts planning costs by 50%PRO
- Quality audit pipeline— auto-scored, auto-fixed, senior-level final reviewPRO
- Tendy chat— AI orchestration partner — routes tasks automaticallyPRO
- GitHub integration— clone, branch, push, open PRsPRO
- Live preview & annotation— see your app render, draw feedbackPRO
- Analytics dashboard— cost breakdown, quality scores, completion statsPRO
- Execution levers— pattern extraction, error memory, scope pre-read, rich specsPRO
- Priority support
Why Pro pays for itself
Real savings, measured across 250+ trials
The Knowledge Graph, quality audit pipeline, and execution levers compound savings every session. By month 2, Pro costs less than the LLM savings it generates.
+30%Higher quality vs raw OpusMeasured in a 90-run A/B study: Tendril’s pipeline scored 30% higher on quality overall than a single Claude Opus agent working the same task (73.7 vs 56.6, p=0.012). Comparison is against using an LLM by itself — not a bare pipeline.
+77%Bigger wins on harder workOn the hardest task in the study (password reset with edge cases), Tendril scored 77% higher than a single Opus agent — 49.5 vs 27.9. The gap grows as tasks get harder.
KGKnowledge Graph compoundingPro agents are primed with patterns, memory, and scoped context from past sessions — so planning starts where the last run left off instead of from scratch.
Compare Plans
Feature
Free
$0/mo
Pro
$29/mo
Bring Your Own API Key (BYOK)
Full pipeline (discover, plan, execute)
Diff viewer & code review
Security monitoring & quarantine
Auto-verify on edits
MCP server support
Active projects
1
Up to 25
Parallel sub-agents
2
Unlimited (hardware-limited)
Knowledge Graph
Quality audit pipeline
Tendy chat (AI orchestration)
GitHub integration (clone, push, PRs)
Live preview & annotation
Analytics dashboard
Execution levers (patterns, error memory)
Priority support
Frequently Asked Questions
Everything you need to know before getting started.
What is BYOK and how do I get started?
What does the free tier include?
How does the Knowledge Graph work?
Can I use my Claude Code subscription instead of a raw API key on Pro?
Can I cancel Pro at any time?
Is my code ever sent to Tendril's servers?
What's the difference between Free and Pro for running multiple projects?
Install free — no credit card required. Add your own API key and start building.