bastion-trust-runtime
Integrate Bastion programmable policy enforcement with trustless-ai agents. Compose ERC-8004 identity + ERC-8263 anchor proofs + ERC-8281 OCP commitments with Bastion's policy engine and human-in-the-loop gate. Use when building agents that need enforceable trust rules, multi-chain execution, or recompute-able audit trails.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-13
- Stars
- ★ 3
- Forks
- 1
01 前言 Front Matter
- name
- bastion-trust-runtime
- description
- Integrate Bastion programmable policy enforcement with trustless-ai agents. Compose ERC-8004 identity + ERC-8263 anchor proofs + ERC-8281 OCP commitments with Bastion's policy engine and human-in-the-loop gate. Use when building agents that need enforceable trust rules, multi-chain execution, or recompute-able audit trails.
03 SKILL.md 正文
Bastion Trust Runtime - Agent Skill
This skill teaches agents to compose Bastion's programmable trust layer with the trustless-ai ERC stack.
What Bastion Adds
trustless-ai provides verification (ERC-8263 anchor proofs, ERC-8274 proof verification, ERC-8281 commitments). Bastion adds the layer between them:
ERC-8004 (identity) → Bastion Policy Engine → ERC-8301 (dispatch) → ERC-8281 (provenance) → ERC-8299 (WYRIWE) → ERC-8274 (verify) → ERC-8263 (anchor)
| trustless-ai gives you | Bastion adds |
|---|---|
| ERC-8004 agent identity | Programmable policy evaluation (11 rule types) |
| ERC-8263 anchor proofs | Pre-execution simulation (Solana + EVM) |
| ERC-8281 OCP commitments | Human-in-the-loop override gate |
| ERC-8299 WYRIWE provenance | Multi-chain TrustAdapter (8 chains) |
| ERC-8275 settlement | Web2 API policy enforcement |
Architecture
Agent Framework (LangGraph / CrewAI / OpenAI Agents SDK)
│
▼
trustless-ai agent-sdk (ERC clients + recompute)
│
▼
Bastion Trust Runtime
├── Identity (ERC-8004 compat)
├── Policy Engine (11 rules)
├── Simulation (Helius / eth_call)
├── HITL Gate (/override)
└── Audit (Anchor PDA + EIP-712 + ERC-8263)
│
▼
Execution Environments (Ethereum, zkSync, Solana, Base, ...)
TrustIntent: Declare What, Not How
Instead of calling low-level ERC methods directly, agents declare a TrustIntent:
intent: transfer
asset: USDC
amount: 500
recipient: 0x...
requirements:
- humanApproval
- sanctionsCheck
- maxRisk: medium
- settlement: ethereum
Bastion resolves:
- Which
IPolicyRuleinstances to evaluate - Whether
PendingHITLis required - Which chain adapter to use
- Whether to anchor via ERC-8263
Using Bastion with agent-sdk
// 1. Register identity via ERC-8004 (trustless-ai)
import { IdentityRegistryClient } from "@trustless-ai/agent-sdk/identity/ERC8004";
const identity = new IdentityRegistryClient({ rpcUrl, address: registryAddress }, account);
const { agentId } = await identity.register("ipfs://Qm...");
// 2. Simulate + policy-check via Bastion (programmable trust)
import { BastionSidecar } from "@zkos-labs/bastion-agentique";
const sidecar = new BastionSidecar({ baseUrl: "https://bastion-agentique.fly.dev" });
const decision = await sidecar.simulate({ transaction, intent: "swap 1 SOL to USDC" });
// 3. If passed, execute. Bastion writes an audit record with ERC-8263 AnchorProof.
// The audit ID can be recomputed offline using verify.recomputeObservationDigest().
// 4. Recompute verification (trustless-ai philosophy)
import { verify } from "@zkos-labs/bastion-agentique";
const result = verify.verifyAuditRecord({
rawInput: JSON.stringify(transaction),
decision: "Pass",
payloadHash: txHash,
expectedObservationDigest: auditRecord.observation_digest,
expectedWyriweHash: auditRecord.wyriwe_hash,
});
// result.valid === true → no trust in Bastion required
Recompute Guarantee
Every Bastion audit record carries:
observation_digest(ERC-8281 OCP): sha256 of "Decision:payload_hash"wyriwe_hash(ERC-8299 WYRIWE): triple-hash binding raw input → sanitized input
Both are recompute-able from public inputs. The @zkos-labs/bastion-agentique verify module provides pure functions that reproduce these hashes without any network access - implementing trustless-ai's "Don't trust. Recompute." guarantee for the policy layer.
When to Use Bastion
- You need enforceable rules (not just advisory): amount caps, allowlists, rate limits
- You need human approval gates for high-value or suspicious actions
- You need multi-chain execution (Ethereum + Solana + zkSync + Base from one interface)
- You need recompute-able audit trails that external parties can verify independently
- You need Web2 API policy enforcement alongside on-chain policy (OpenAI, Stripe, Slack)
04 同仓库更多 Skills
bastion
Programmable Trust Runtime for Autonomous Systems. Orchestrates identity, policy, privacy, durable execution, and multi-chain settlement.
web2-api-firewall-administrator
Guides configuring Bastion Web2 API adapter policies for the Trust Runtime. Covers endpoint allowlists, provider budgets, rate limits, content inspection (PII/secrets detection), header filters, and OpenAPI spec-based auto-configuration. Use when configuring API call security for AI agents, setting provider spending caps, inspecting outbound API traffic, or setting up the Web2 adapter.
transaction-screening-workflow-concepts
Educational map of transaction-centric compliance screening-transfer as the atomic unit, deposit vs withdrawal direction, single and CSV import, transaction list and detail views, per-transfer screening, rescreen, and STR-style exports. Use when the user asks how monitoring UIs treat tx hashes, directions, or regulatory reporting hooks-not for legal filing advice or evading reporting.
solana-tracing-specialist
Guides Solana-specific on-chain forensics-ATA resolution, SPL instruction parsing, transaction history via RPC and indexers (e.g. Helius-style APIs), fund-flow graphs, Solana clustering heuristics, and program authority review. Use when the user investigates Solana wallets, SPL tokens, DEX/Jito flows, rug or phishing patterns on Solana, or needs evidence-structured tracing reports with public data only.
solana-onchain-intelligence-resources
Companion to blockint-skills Solana work-official doc indexes and stacks for RPC/indexers, sanctions and address intel MCP, web research, agent payments, graph UIs, Solana Foundation dev skills, Solana Policy Institute (policy/regulatory education), optional open-source Scrapy collection (BlockchainSpider), Impersonator Solana (dApp dev/testing), coral-xyz sealevel-attacks (security pattern examples), Neodyme Solana Security Workshop (workshop.neodyme.io), and Osec Solana auditor introduction blog (osec.io). Use when the user needs pointers to Helius, Range AI, Tavily, PayAI x402, React Flow, Solana.com skills, or SPI alongside on-chain forensics-not as a substitute for reading current provider docs, legal counsel, or API keys.
solana-defi-vulnerability-analyst-agent
Guides discovery and documentation of Solana DeFi protocol risks from public code and chain state-Anchor/native programs, PDAs, CPIs, oracles, pools, SPL mechanics, and historical tx reconstruction. Use when the user asks for Solana program security review, DeFi vulnerability triage, PDA or CPI safety, oracle or liquidity-pool risk, launchpad/bonding-curve issues, or evidence-backed severity findings without exploits or private keys.