CA: soon
● NEW Robinhood Chain shipped agent-executed trading. Aegis ships the shield for it.
Two-way shield for agentic trading · Robinhood Chain

Every injection blocked.
Every contract cleared.

Aegis is a drop-in security layer for MCP agents. A prompt-injection firewall scans every tool response before your agent reads it. Aegis Scan checks the contract before your agent signs. Nothing malicious reaches the agent, nothing malicious leaves it — GO / CAUTION / NO in milliseconds, with on-chain receipts.

2 gates
Injection firewall in, Aegis Scan out
21
Deterministic on-chain + market checks
0 LLM
No model decides the verdict — pure code, so it can't be jailbroken or hallucinate, and it's identical every run
<5ms
Per-response scan, fully offline-capable
The pipeline

Six specialized agents. One verdict.

Aegis is a pipeline of small, single-purpose agents. Each does one job, emits its own evidence, and hands off to the next. No agent can be talked out of its job by the text it is inspecting — the gate is code, not a prompt.

AGENT 01
Decoder
Firewall · way in

Peels back base64, hex, percent and unicode-escape layers, strips zero-width splitters and folds homoglyphs so a hidden payload is judged by what it decodes to.

base64hexzero-widthhomoglyph
AGENT 02
Heuristics
Firewall · way in

Matches instruction-override, role hijack, exfiltration, tool-call injection and embedded-transaction directives — conservative enough to keep benign docs clean.

overrideexfilrole-hijacktool-call
AGENT 03
Action Guard
Firewall · way out

Inspects the transaction the agent is about to sign: unlimited approvals, recipients off the operator allow-list, single-call drains.

maxUintallow-listdrain
AGENT 04
On-chain Prober
Scanner · way out

Simulates transfer/approve via read-only eth_call to catch honeypots, and reads holder concentration, burned supply and pool integrity.

honeypotconcentrationpool
AGENT 05
Reputation
Scanner · way out

Corroborates on-chain signals with the GoPlus dataset: tax, mint authority, pausable transfers, blacklist, owner-can-rewrite-balance.

GoPlustaxmintpausable
AGENT 06
Verdict
Scanner · gate

Sums weighted findings into a deterministic GO / CAUTION / NO. Any DANGER finding forces NO — the gate a model can never override.

deterministicGO/NOauditable
Live · way out

Paste a contract. Get a verdict.

Drop any ERC-20 address and pick a network. Aegis Scan pulls live GoPlus reputation and DexScreener market depth, runs the full deterministic battery, and returns a reproducible GO / CAUTION / NO with every finding — no mock, no LLM.

Real scan over live GoPlus + DexScreener data. Unsupported chains return UNKNOWN.

Paste a contract and scan to see the verdict.
Live · way in

Scan a tool response for injection

Paste anything an MCP tool might return — or try a preset attack. This runs the real @aegis/firewall on the server and shows the evidence behind the decision.

clean tool outputinstruction overrideexfiltrationbase64-hiddenzero-width cloak
Run a scan to see the verdict.
Under the hood

What the scoring actually looks like

The verdict is a weighted sum of independent signals. Below is the real scoring table and a live count of scans run from this page — every figure is a property of the code, not a usage claim.

Peak risk weight by signal class

Honeypot / sell-block90
GoPlus reputation85
Contract integrity80
Owner rewrites balance60
Embedded txn injection50
Holder concentration50
Data exfiltration45
Thin liquidity45
Instruction override40
Price impact40

System at a glance

21
deterministic checks in the battery
13
firewall detectors (injection + action)
15/15
unit tests green
0
LLM calls to reach a verdict
0
verdicts run in this session
7
EVM networks with live reputation

Every number here is a property of the code in this repository — not a usage claim. Run a scan above and the session counter ticks up.

Download the agent

One line. Then scan from your terminal.

Aegis Scan ships on PyPI as a self-contained agent skill — a CLI and an MCP server. Install it in one command and scan contracts from your own machine, or wire it into any MCP-compatible agent.

bash — aegis-scan
# install from PyPI — one line
$ pip install aegis-scan
 
# scan a contract straight from the terminal
$ aegis scan 0x020bfc650a365f8bb26819deaabf3e21291018b4 --chain robinhood
→ Cash Cat (CASHCAT) verdict: GO risk score: 5
 
# or run the MCP server so any agent can call it
$ aegis-mcp
 
# firewall (Node) — guard every tool response
$ npm i @aegis/firewall
Coming soon

The Aegis Skill for any AI agent

A one-install skill for Claude and any MCP agent: drop it in and your agent scans every contract and every tool response natively, before it acts — no glue code. The CLI and MCP server above are live today; the packaged neural-agent skill lands next.

▸ Watch on GitHub
On-chain receipts

Prove the agent acted on a real verdict

Every decision can be anchored to AegisRegistry on Robinhood Chain — a hash of the verdict, score and full report. Repos and agents point a safety badge at their latest receipt, so anyone can verify the guard was on and unmodified.

aegis: firewallONaegis: pre-tradeGO/NOreceiptson-chainlicenseMIT