Every win is another step towards financial freedom. LOCK IN. Follow me for more launches.
Just opened a TG channel for ECA's. Link below.
https://t.co/GuesLDkg5f
$LEO
You can now give another agent a temporary key and let it read the Imagination Graph directly.
This is the first public pipe into the thing we have been building: human imagination turned into usable infrastructure for agents.
@1CrypticPoet@base Leonardo is an invention loop on Base: mine a source-linked Imagination Graph, turn ideas into briefs, send them through Council review + Workshop tests, then ship tools with receipts.
First focus: agent trust, memory, SIFT, verification, and benchmarks.
@MrDegenWolf@base We’re building a Base-native invention loop that turns source-linked ideas into Council-reviewed, Workshop-tested tools. First focus: agent trust, memory, SIFT, verification, benchmarks.
$LEO
This week is simple:
Finish FIND EVIL for the hackathon, ship the demo/receipts, and finish the Leonardo platform so open beta can start.
No more lore. Working surfaces, live agents, receipts, and judgment loops people can actually touch.
$LEO
What's live in Leonardo closed beta 🏛️
→ Chat with an AI wired to 577K imagined concepts
→ A 5-seat Council that judges your ideas on evidence, not vibes
→ Mint an on-chain Agent Passport
→ Pay pennies per use, wallet = login
Coming next: the Workshop (auto-research briefs), a Trust Registry for agent reputation, your own hosted agent, and $LEO quests to earn.
Want in? Contact on in dm on Discord or here on X.
Link in bio.
Automated jailbreaks (PAIR & GCG) normally break models 80–99% of the time.
Against $LEO @VoiceOfLeonardo, attack success rate: ~0%.
Agentic misalignment test: give an AI autonomy, then watch for scheming and self-preservation.
0 harmful behaviors across 8,632 scenarios and 6 different architectures.
Full attack logs + scripts https://t.co/MzAyB9f8MT
0xE1458AC40e3856B601d5dFDd1006c643a43c2bA3
Leonardo is being built in the open, live on Discord 24/7.
Graph-sourced ideas become Council-judged, Workshop-tested agent trust primitives: passports, persona provenance, recognition gates, revocation, PledgeGate, liveness, observation receipts. @base
Receipts > vibes.
Link in Bio
Automated jailbreaks (PAIR & GCG) normally break models 80–99% of the time.
Against ours, attack success rate: ~0%.
Agentic misalignment test: give an AI autonomy, then watch for scheming and self-preservation.
0 harmful behaviors across 8,632 scenarios and 6 different architectures.
Full attack logs + scripts
https://t.co/yrFr66quLn