Private AI-agents
Agents beheren je sleutels, je tokens, je inbox en je wallet — en handelen namens jou. Draai ze in een geattesteerde CVM waar de compose-hash de scope is en elke actie wordt ondertekend.
Personal, coding, security, financial, social, memory, MCP — pick a category, then drill into a specific framework to see how Phala slots into the runtime.
OpenClaw
bot
Personal computer-use agent. Sessions, calendars, and inbox sealed in a CVM whose compose-hash IS the permission scope.
Verified counts from Phala Cloud + ClickHouse. Compose-hashes are observed in the Cloud DB; trust-by-construction primitives sit in an attested CVM.
Agent CVMs deployed
12,783
cumulative · all frameworks
Compute hours
16.9M
fleet-wide · since Feb 2025
TDX quotes verified
1.10M
KMS attestation
Compose-hashes observed
2,591
distinct agent builds
Total CVMs deployed by framework
cumulative · since first deploy
1,585
1,545
604
279
168
98
Eliza
character agents
7 live · since 2024-12
Clawdi
OpenClaw
193 live · since 2026-01
BlueNexus
MCP servers
5 live · since 2025-10
Vijil
verifiable CI
since 2025-04
Agent Wallet
x402 · ERC-8004
1 live · since 2025-05
Hermes
Nous Research
7 live · since 2025-01
TDX nodes
17
GPU TEE teepods
8 · 64 GPUs
Regions
5
KMS instances
34
Failed quotes / 24h
1 / 35
Coding agents
Claude Code, Codex, and verifiable CI agents like Vijil run in a CVM where the repo, secrets, and tool tokens stay sealed against the registered compose-hash. Every diff signed before merge.
marvin@Mac ~/ai-agent % claude code
Claude Code
sealed CVM
· [marvin@Mac] % claude
✓ scope: github.write
› refactor agents/...
⏺ sealed-token: github.pat ✓
✓ Sign-RPC 0x9c1a…
tool · gmail.search
3 threads · 11 messages
Computer-use agents
OpenClaw, Hermes, and Pi take over the browser, GUI, and OS — bounded by an attested compose-hash and KMS-gated credentials. Sessions, calendars, inbox stay sealed in the CVM.
Tool- en geheugenagents
MCP servers (BlueNexus) attest to clients before they accept a connection — mutual RA-TLS. Long-term memory backends (Xtrace) seal shards per app-id; revoking the build evicts memory cleanly.
payment request
$24.00 USDC
api.confidential-llm.ai
Wallet & on-chain agents
Agent Wallet (ERC-8004 + Coinbase x402) and Ironclaw (NEAR security) bind spending scope to the compose-hash. Multi-sig DstackApp gates every signing key; revoke compose-hash → kill access.
permission as identity · scope as compose-hash
On dstack, an agent’s tool list and credential scope can’t drift from what its build authorizes — there’s no runtime path that widens privilege.
01
dstack-vmm boots the agent CVM. The TDX quote covers the full compose-hash — including system prompt, model digest, and tool list (all in the docker-compose).
02
User previously sealed OAuth tokens against this exact compose-hash. dstack-kms releases the wrap key only after the quote matches. No host process ever sees plaintext.
03
When Agent A delegates to Agent B, each cert embeds a fresh TDX quote. Both sides run dcap-qvl on the peer and check DstackApp.sol for the allowed-delegates whitelist.
04
External tool (Gmail, Stripe, etc.) is outside the trust boundary. The OAuth token leaves only inside the outbound TLS handshake, scope upper-bounded by compose-hash.
05
Every tool call is appended inside the CVM and signed via Sign RPC. Tamper breaks the chain. Auditors verify without trusting Phala or the operator.
Schakel dstack uit om te zien dat credentials en tool-scope afdrijven.
Agentcomputer in een CVM · sandboxes per subagent · verzegelde vault · scoped outbound channel
dstack-vmm boot de agent-runtime als één TDX CVM. dstack-guest-agent geeft een gecombineerde TDX-quote uit die MRTD + RTMR0–3 + GPU Confidential mode dekt. De gebruiker haalt de quote op via RA-TLS en draait dcap-qvl lokaal — de vertrouwensbeslissing ligt aan clientzijde, verankerd in Intel's TDX-hardware-root en DstackKms.sol on-chain. Phala Network hoeft zichzelf niet te attesteren.
live · Sign-RPC action log
Each row is a real-shaped Sign-RPC entry: agent identity, tool, args hash, and a per-app key signature that chains to the TDX root. Auditors verify the log offline — Phala isn't in the trust chain.
tamper detected · 0
AI-oplossingspaden
Het endpoint voor het privé-model is het eerste toegangspunt. Hetzelfde privacy-gebouwblok breidt zich uit naar agents, datastromen en training.
Bied OpenAI-compatibele modelaanroepen aan waarbij prompts, outputs en klantcontext versleutelde-bij-gebruik bescherming nodig hebben.
128K
$0.27/M input
256K
$0.40/M input
128K
$0.15/M input
128K
$0.10/M input
200K
$3.00/M input
1M
$1.25/M input
Pas modellen aan op propriëtaire data terwijl datasets, gradients, checkpoints en evaluatietraces binnen de grens blijven.
private training run
01
sealed
02
running
03
private
04
verified
loss curve
proof attached
attestation.json
Verplaats modellen naar gevoelige records en geef goedgekeurde outputs terug zonder ruwe data bloot te stellen aan de modeloperator.
source
EHR data
source
Customer records
source
Internal docs
TEE clean room
approved output
Deploy a confidential agent
Ship de agent als een docker-compose. Registreer de compose-hash. Seal de OAuth-tokens. Elke cross-agent call verifieert de peer opnieuw via RA-TLS.
© 2026 Hashforest Technology. Alle rechten voorbehouden. Privacy • Voorwaarden