Who approved the action?
Agent work needs explicit capabilities and policy boundaries instead of open-ended access.
Utility-first · Base network
The utility layer for autonomous agent work.
AI agents are becoming capable of acting on their own. The hard part is no longer execution. It is deciding what they may do, how much they may consume, and how to prove what actually happened.
FLC gives distributed AI agents a shared utility layer for budgeting work, bounding authority, measuring execution, and producing verifiable receipts.
A1Research
A2Operations
A3On-chain
A4Services
FLCbudget · execution · proofdistributed agents→bounded work→verifiable receipts
NetworkBase
SymbolFLC
Official contract0x320f7aaeefec65f29b7e4509739ff9c62ded9803
MigrationEthereum → Base
01 · Why FLCAI agents can act.
AI agents can act.
But who controls the work?
Autonomous software should not receive unlimited authority. Every useful action creates three questions that must be answered before agent networks can operate safely at scale.
How much authority did it have?
Each task needs a measurable utility budget, retry policy, and duration limit.
How do we prove what happened?
The final state must connect intent, external effect, verification, cost, and evidence.
FLC turns those questions into machine-readable constraints around approved work.
02 · Product model
From intent to proof.
A distributed agent does not receive a blank check. Work moves through a bounded lifecycle, and each stage adds evidence that can be inspected by software or people.
01Intent
A specific task is requested.
→
02Budget
Authority and utility are bounded.
→
03Action
An approved capability executes.
→
04Verify
The external effect is read back.
→
05Proof
Receipt, cost, evidence, and state become auditable.
intent → budget → bounded action → external effect → deterministic verification → receipt03 · CapabilitiesOne utility model.
One utility model.
Many kinds of agents.
FLC is capability-agnostic. Research, operations, commerce, machine services, and on-chain execution can share the same bounded-work model while keeping their own policies and verification rules.
Acquire evidence, refresh sources, continue bounded investigations, and return traceable research output.
Run approved workflows, generate deliverables, call stable application capabilities, and record completion.
Meter task-level access to compute, models, tools, or other machine-readable services.
Run allowlisted contract actions, simulate before execution, and verify transaction plus post-action state.
DeFi is optional. Swaps and liquidity operations are examples of bounded action modules, not the definition of FLC.
04 · Safety
Autonomy without unlimited authority.
Agent utility only works when execution remains controlled, observable, and accountable. Safety is part of the execution model—not an afterthought.
FLC · Flowchain
Utility moved up the stack.
FLC began as utility for Flowchain's private blockchain and device-oriented ecosystem. Its current role is higher-level: accounting for bounded execution and verifiable work across distributed AI agents.
Utility, not an investment promise. This site describes product utility and technical architecture. It does not offer a token sale, promise returns, or provide financial advice. Flowchain support will never ask for your seed phrase or private key.