The verifiable guardrail layer forautonomousAIagentsonSolana

One policy. Any program. An agent’s action is checked against a declarative policy, that decision is proven in zero-knowledge, and the proof is verified on-chain — before the action executes.

includes GLYPH agent skillnpx skills compatibleagent-agnostic repo guide
Same proofs
one RISC Zero circuit
Same policy engine
one declarative DSL
Same agent skill
one repo operating model
Different programs
System · Token · Memo · …
intent/policy/commitment/journal/proof/pairing/execute/intent/policy/commitment/journal/proof/pairing/execute/
Based on arXiv:2509.00085

A concrete implementation of Tobin South’s AI trust thesis.

GLYPH is a working Solana implementation inspired by the arXiv thesis Private, Verifiable, and Auditable AI Systems by Tobin South. The paper argues for AI systems that combine privacy, verifiability, auditability, confidential deployment, and scoped delegation; GLYPH instantiates that architecture for autonomous agents through BYOK policy compilation, TEE-side checks, RISC Zero proofs, and Groth16 enforcement.

Read arXiv:2509.00085 ↗live Solana implementationagent-agnostic guardrails
01

Delegation

human intent binds agent authority to a concrete policy commitment

02

Privacy

policy evaluation can remain inside a TEE while exposing only commitments

03

Verifiability

RISC Zero execution produces a journal that is proven and checked on-chain

04

Auditability

Solana records the verifier state, policy commitment, and replay-safe action trail

The problem

Agents are moving real value with no enforced limits and no proof they stayed in bounds.

Autonomous agents now hold keys and sign transactions on Solana. Today the only thing standing between an agent and your funds is the agent's own code — a trust black box. If it misbehaves, gets jailbroken, or is simply wrong, there is no on-chain limit and no evidence of what it was allowed to do.

!

No enforced limits

Spend caps, allowlists and rate limits live in off-chain code an attacker can bypass. The chain accepts whatever the key signs.

!

No proof of compliance

Even a well-behaved agent can't prove it stayed within policy. There is nothing auditable to point to after the fact.

!

Trust doesn't compose

Every protocol re-invents its own bot safety. Nothing is portable across perps, prediction markets, or privacy apps.

GLYPH is a horizontal trust layer. It doesn’t compete with perps, prediction or privacy protocols — it makes agents safe across all of them.

Same proofs, same policy engine — different target programs.

End-to-end

From intent to on-chain enforcement, in four steps

Walk the full pipeline. Each step produces a cryptographic artifact the next step consumes — culminating in a proof Solana verifies before anything executes.

Step 1 · Agent

Build a signed TransactionIntent

The agent assembles its desired action — target_program, accounts and data as opaque bytes — and signs it with Ed25519. GLYPH never needs to understand the target program.

signed intent

TransactionIntent { target_program, accounts[], data, sig }

Live system surface

Thirty moving parts, one enforceable path

The demo surface mirrors the product architecture: identity, policy compilation, deterministic hashing, ZK proving, and on-chain enforcement are visible as one connected system.

intent/policy/commitment/journal/proof/pairing/execute/intent/policy/commitment/journal/proof/pairing/execute/
01BYOK LLM proxy
02Schema guard
039-rule DSL
04Policy TOML
05Wallet identity
06Agent registry
07Ed25519 delegation
08Intent hashing
09Tx binding
10Nonce PDA
11TEE attestation
12Policy commitment
13RISC Zero guest
14Groth16 wrap
15BN254 pairing
16Journal binding
17VK PDA
18VK rotation
19Devnet deploy
20Live RPC reads
21Lean proofs
22Rust parity
23TS parity
24CI proof run
25x86 prover
26Artifact upload
27Audit trail
28Daily volume
29Time window
30Program allowlist
Live · bring your own LLM

Author a policy in plain English

GLYPH's natural-language → policy DSL compiler, live. Describe what your agent may do; your own LLM (any OpenAI-compatible provider — OpenAI, Anthropic, Gemini, Groq, OpenRouter, xAI, Ollama, LM Studio, or a self-hosted LiteLLM proxy) compiles it to the canonical 9-rule policy. A deterministic schema guard clamps the output, then the real policy_commitment is computed in your browser — byte-for-byte identical to the Rust SDK — and you test intents against it. Connect a wallet to bind the policy to your own agent identity.

Describe the policy
↑ set your LLM key first

This is GLYPH’s natural-language → policy DSL compiler. The LLM proposes JSON; a deterministic schema guard in your browser clamps it to the 9-rule DSL. The agent that writes policy is exactly what GLYPH guards.

compiled policyawaiting input

Describe a policy on the left and compile it. The TOML, the real policy_commitment, and live ALLOW/DENY tests appear here.

Interactive demo

One policy, any program

Pick an agent action. GLYPH evaluates it against a single declarative policy and computes the canonical policy_commitment — live, in your browser, byte-for-byte identical to the Rust SDK, the TEE worker and the on-chain verifier. Three different programs, one identical commitment, and a correctly-denied fourth.

Pick an action4 intents · 1 policy
The one policy
version
1
max_lamports_per_tx
1 SOL
max_daily_volume
5 SOL
max_accounts_per_tx
16
require_signer
true
allowed_programs
System · Token · Memo
System transfer→ TEE policy engine

Native SOL transfer of 0.5 SOL — under the 1 SOL cap, program allowlisted.

target_program
accounts
2 metas
max_lamports
0.500 SOL
data (opaque bytes)
AgAAAECcXAAAAAAA
policy_commitment — SHA-256, computed in your browser

Same hash for System, SPL Token and Memo. The policy binds the agent — not the target program. That is what makes GLYPH horizontal by construction.

intent_hash
tx_hash → binds ix[1]
Byte-for-byte parity with the Rust canonical serializercomputing…

Program-agnostic by construction. Intents carry target_program, accounts and data as opaque bytes. The allowlist is a per-agent policy field — never hardcoded — so the same engine guards any program without code changes.

Live on-chain

Real infrastructure, deployed to devnet

These accounts are queried live over the public devnet RPC on page load — not screenshots. The verifying key is a real seeded Groth16 VK, and every initialization transaction is confirmed on-chain.

honest status

The final verify_and_execute step requires a Groth16 proof generated on x86 (Apple Silicon can’t produce it locally). The CI proof-generation path now completes; we still don’t claim the full round-trip has executed on-chain until that proof is landed in a devnet transaction. Everything else shown here is live and real.See the deploy tx ↗

Defense in depth

Three independent layers of trust

GLYPH does not ask you to trust one thing. An attacker must defeat hardware attestation, a zero-knowledge proof, and on-chain consensus — all three — to push a non-compliant action.

01SGX · Nitro · SEV

Hardware

Trusted Execution Environment

Policy evaluation runs inside a hardware enclave that attests to the exact code it executed. The host OS can't tamper with the decision.

02STARK → Groth16

Cryptography

RISC Zero zkVM

The decision is re-proven in a zero-knowledge VM. The resulting succinct proof attests that the policy was satisfied — without revealing the policy internals.

03real BN254 pairing

Consensus

On-chain Groth16 verifier

Solana itself verifies the proof with a real BN254 pairing check before the action executes. Enforcement lives in consensus, not in a promise.

Why horizontal beats vertical

A vertical bot competes with 100 bots. A trust layer compounds with the ecosystem.

Every new agent and every new program is a user of GLYPH — not a competitor. Value grows with the network instead of fighting it.

Other approaches
GLYPH
Build an agent
Make all agents safe
Vertical app, one protocol
Horizontal guardrail, any protocol
Trust the bot
Cryptographic proof of compliance
Off-chain promises
On-chain enforcement
Research credibility

Research paper implemented as live Solana infrastructure

GLYPH implements the layered accountability architecture from Tobin South’s MIT PhD dissertation, “Private, Verifiable, and Auditable AI Systems.” The paper is a PhD thesis in cryptography and security that combines zero-knowledge verifiable claims, confidential deployment with TEEs/MPC, and scoped delegation for autonomous and multi-agent AI. This interface is the implementation layer: natural-language scope, authenticated delegation, confidential enforcement, succinct proof, and auditable on-chain execution.

P
Private
policy internals never leave the enclave / proof
V
Verifiable
succinct ZK proof of policy satisfaction
A
Auditable
on-chain commitment + nonce trail per action

19 Lean 4 theorems · 0 sorry — the policy semantics are formally verified.

Agent skill

GLYPH ships with a committed agent skill

The repo includes a project-scoped glyph skill for coding agents. It is part of the public repository, discoverable with npx skills, and tells any compatible agent how to work on GLYPH without weakening the policy, proof, verifier, wallet delegation, or arXiv:2509.00085 alignment guarantees.

npx skills add https://github.com/guglxni/glyph --skill glyph
npx skills compatibleagent-agnostic instructionsresearch-aware guardrails
.agents/skills/glyph/SKILL.md
Global discovery
installable from GitHub with npx skills add https://github.com/guglxni/glyph --skill glyph
Thesis map
keeps agents anchored to arXiv:2509.00085 and the trust-stack mapping
Repo map
routes work to the right frontend, circuit, TEE, verifier, and docs paths
Invariant guard
preserves policy commitments, journal bytes, tx binding, VK integrity, and replay safety

Verify locally: npx skills add https://github.com/guglxni/glyph --list finds glyph, and npx skills list --json shows it after install.

0
passing tests
0
Lean 4 theorems · 0 sorry
0
rule policy DSL
0
job CI pipeline
real BN254 pairingRISC Zero zkVMSGX · Nitro · SEVprogram-agnostic by construction

Make every agent safe — cryptographically.

One policy. Any program. Cryptographically proven, on-chain.