AEGIS · Agentic Compliance · Live

Compliance that
checks itself.

AEGIS is not just a monitoring layer. It's an engine that checks, signs, and locks in evidence for 158 controls across 12 regulatory frameworks on its own, with no human hand on the evidence chain. The system builds the bundle your auditor reads.

96.2 / 100
Composite Score · AEGIS Last scanned: 2026-05-08T00:00:00Z
Live
6.5 ms / control
460 ms full scan
158 controls
12 frameworks
2.32x cross-framework leverage
32.9% machine-verified

96.2
/ 100
Composite attestation score across all 12 frameworks
6.5 ms
Per-control audit (target <500 ms; live typical)
460 ms
Full-fleet scan of all 158 controls (target <30 s)
2.32x
Cross-framework leverage: one setup, average 2.32 requirements satisfied
158
controls total
65 SOC 2 TSC 2017 + 93 ISO 27001:2022 Annex A
32.9%
machine-verified
Cryptographic probes with signed evidence envelopes
12
frameworks
SOC 2, ISO 27001/17/18/701/36, 42001, EU AI Act, GDPR, eIDAS 2.0, NIS2, DORA

The Thesis

Watching your systems isn't the same as proving they're compliant

Most compliance tools scan your infrastructure and watch for problems. They flag things that look wrong. But they don't sign anything, lock anything in, or create proof you can check that shows the control's real state at a specific moment. That gap between watching and proving is what makes audits expensive.

The monitoring approach

Ongoing monitoring
Tools like Vanta, Drata, and Secureframe scan cloud environments now and then for problems like missing MFA, open ports, or unencrypted storage. The result is just a log of what they saw.
A human writes the report
For a real SOC 2 opinion, a CPA firm reviews the monitoring logs and talks to your staff. A human auditor puts together the evidence that actually matters, not the system.
Evidence isn't signed
Monitoring logs just record what the tool saw. There's no cryptographic proof tying that observation to the real control state at that moment.
Framework coverage is narrow
Most tools handle SOC 2 and ISO 27001 out of the box. Mapping to GDPR, NIS2, DORA, EU AI Act, or eIDAS 2.0 takes manual setup and someone who knows those rules well.
Overlap between frameworks is manual
When you need to satisfy two frameworks at once, the control mapping is a spreadsheet job a consultant does by hand. There's no built-in way to reuse work across frameworks.

The proof-based approach

Ongoing proof, not just watching
AEGIS doesn't just observe. It checks each control through the POST /v1/aegis/control/audit loop, captures the result, and signs the evidence envelope with Ed25519 and ML-DSA-65 in the same pass.
AEGIS builds the evidence bundle itself
The output of POST /v1/aegis/scan is a bundle your auditor can use right away. AEGIS-Report packages it for export. The auditor reads what the machine signed, not what a person put together.
Evidence that's locked with math
Every envelope carries a SHA3-256 digest, an Ed25519 signature, and a post-quantum ML-DSA-65 signature. AEGIS-Seal locks each bundle onto Base L2. Nobody can quietly tamper with the evidence and get away with it.
12 frameworks from one setup
HIVECOMPLY maps each control you set up to every framework requirement it satisfies. The cross-framework leverage is 2.32x. One setup covers an average of 2.32 regulatory requirements at once.
Built for the post-quantum world, tied to real-world randomness
Signatures come from the TACHYON family (ML-DSA-65, SLH-DSA). The randomness used to make keys comes from MAPET-X: 24 physical sources across 6 groups, including soil microbial activity (Loess) and cosmic ray flux (FERMION).

AEGIS-Core

The self-check loop

Each control goes through a five-stage pipeline. It runs on STM32N6 (Arm Cortex-M33, hardware AES-256). It typically takes 6.5 ms from starting the check to having a signed envelope ready to export.

What this means for your auditor

When your auditor reviews AEGIS evidence, they're reading outputs the system signed, not notes a person put together from screenshots. The SHA3-256 digest ties the evidence to the control's real state at the moment it was checked. The Ed25519 and ML-DSA-65 signatures prove the digest hasn't changed since the system created it. The Base L2 anchor stamps a timestamp on-chain that nobody can fake or backdate. No part of the evidence chain depends on a person holding onto it.

Sample evidence envelope: CC6.1

JSON · POST /v1/aegis/control/audit
{
  "control_id": "CC6.1",
  "pass":       true,
  "score":      1.0,
  "duration_ms":6.53,
  "sha3_256":   "99f432dd70f9cd54a3b2e1078fdf44c2a73b...",
  "ed25519":    "3O7pN2XqR1sK9vA...base64-truncated...",
  "ml_dsa_65":  "MIIBIjANBgkqh...base64-truncated...",
  "anchored_at":"2026-05-08T00:00:00Z",
  "chain":      "base",
  "block":      18447293
}

Competitive Landscape

Watching all the time vs. proving all the time

Vanta, Drata, and Secureframe are monitoring tools. They watch your infrastructure and build dashboards showing where you stand. AEGIS is a proof engine. This isn't just a marketing difference. It changes what evidence your auditor can actually rely on.

Capability Vanta / Drata / Secureframe Hive AEGIS / HIVECOMPLY
Approach Continuous monitoring
(infrastructure observation)
Continuous attestation
(system signs its own evidence)
Evidence format Compliance posture dashboard; screenshots; human-assembled logs Machine-signed JSON envelopes with SHA3-256 + Ed25519 + ML-DSA-65
SOC 2 report production Human CPA firm produces the report; tool supplies supporting data AEGIS-Report produces the auditor-ready evidence bundle; CPA reviews machine output
Cryptographic signatures on evidence None. Logs are unsigned records Ed25519 + ML-DSA-65 on every envelope; SHA3-256 integrity digest
Post-quantum readiness Not offered TACHYON ML-DSA-65 (FIPS 204) + SLH-DSA on every signed artifact
On-chain anchoring Not offered AEGIS-Seal anchors bundles on Base L2 with tamper-evident timestamp
Frameworks covered SOC 2, ISO 27001, HIPAA, GDPR (broad but shallow) 12 frameworks natively: SOC 2, ISO 27001/17/18/701/36, 42001, EU AI Act, GDPR, eIDAS 2.0, NIS2, DORA
Cross-framework leverage Manual consultant spreadsheet 2.32x built in. One control setup satisfies an average of 2.32 framework requirements
Speed per control Periodic scans (hourly or daily); no real-time audit capability 6.5 ms per control via POST /v1/aegis/control/audit
Full-fleet scan time Hours for a full evidence collection pass 460 ms for all 158 controls
Physics-based entropy source Not applicable MAPET-X: 24-axis physics entropy (terrestrial, quantum optic, subatomic, bio, cosmic, agricultural)
Cryptographic deletion proofs Not offered Signed deletion envelopes via AEGIS-Report; auditable proof that data was erased
Swarm-validated entropy provenance Not applicable T-THRESH consensus across distributed nodes before entropy is admitted to key generation
Self-verify with curl Proprietary dashboard only; no open inspection endpoint curl -s https://receipts.thehiveryiq.com/v1/aegis/score

Vanta, Drata, and Secureframe are effective tools that serve many organizations well. This comparison just points out the structural differences in how each approach builds evidence, not general product quality. Both approaches still need a CPA firm for a formal SOC 2 Type 1 or Type 2 opinion.


Architecture

Seven AEGIS components

AEGIS is made of seven separate pieces, each with its own job. They share a common audit bus and a signing key built from MAPET-X randomness.

AEGIS-Core

Self-check loop

The main check runner. It takes a control ID through POST /v1/aegis/control/audit, runs the check against your live system, captures the pass or fail result, computes the SHA3-256 digest, and hands the envelope to the signing chain. Target speed: 6.5 ms typical, under 500 ms at the outside.

AEGIS-Scan

Full-fleet check runner

Runs all 158 controls in one coordinated pass via POST /v1/aegis/scan. It runs controls at the same time when it can, and in order when the order matters. The full run finishes in 460 ms. Results get rolled up into a framework-level matrix.

AEGIS-Score

Live framework rollup

Works out the weighted composite score across all frameworks. You can pull it via GET /v1/aegis/score. The score reflects the current machine-checked and self-attested control states. Current composite: 96.2. 32.9% of controls are machine-checked.

AEGIS-Report

Bundle your auditor can use right away

Packages signed envelopes for export through HIVECOMPLY (POST /v1/hivecomply/bundle/export and POST /v1/hivecomply/bundle/verify). The output is a JSON archive with every control result, every signature, and the MAPET-X randomness proof. The auditor can check the bundle without touching your internal systems.

AEGIS-Bridge

Connects to outside GRC tools

Translates back and forth with outside governance, risk, and compliance platforms. It takes in outside compliance events via POST /v1/hivecomply/event/ingest. It exports in formats that work with common GRC toolchains. You don't have to rip out your existing GRC tools to use it.

AEGIS-Forecast

Score trend by framework

Tracks how each framework's score moves over time. It flags controls trending toward failure before they cross the line. You can pull it via GET /v1/aegis/matrix and the HIVECOMPLY framework map. It gives your compliance team a heads-up instead of a surprise after the fact.

AEGIS-Seal

On-chain locked proof

Locks each signed evidence bundle onto Base L2 (EIP-4844 compatible). The on-chain record stamps a timestamp that nobody can backdate. The transaction hash is included in the evidence bundle, so auditors have their own way to check it through the public Base block explorer. Treasury: 0x15184Bf50B3d3F52b60434f8942b7D52F2eB436E.

HIVECOMPLY

One dashboard for 12 frameworks

The cross-framework layer that ties it all together. It maps each control to every regulation it satisfies via GET /v1/hivecomply/framework/map and GET /v1/hivecomply/control/status. It works out the 2.32x leverage ratio. It tracks SOC 2 TSC 2017, ISO 27001:2022, ISO 27017, 27018, 27701, 27036, 42001, EU AI Act, GDPR, eIDAS 2.0, NIS2, and DORA in one place.


HIVECOMPLY

Twelve frameworks, one setup

HIVECOMPLY maps every control you set up to every regulation it satisfies. The cross-framework leverage ratio is 2.32x. One setup satisfies an average of 2.32 framework requirements. Total mapped controls: 158.

soc2
SOC 2 TSC 2017
65 controls · AICPA Trust Services Criteria
iso27001
ISO 27001:2022
93 controls · Annex A, 2022 edition
iso27017
ISO 27017:2015
Cloud-specific security controls
iso27018
ISO 27018:2019
PII in public cloud; privacy controls
iso27701
ISO 27701:2019
Privacy information management
iso27036
ISO 27036
Supplier relationship security
iso42001
ISO 42001:2023
AI management system
eu_ai_act
EU AI Act
High-risk AI system requirements
gdpr
GDPR
Data protection; Articles 25, 32, 35
eidas2
eIDAS 2.0
Electronic identification and trust services
nis2
NIS2
EU network and information security
dora
DORA
Digital operational resilience for finance
2.32x
Cross-framework leverage
Set up a control once and satisfy an average of 2.32 regulatory requirements. This is worked out across 158 controls and 12 frameworks via GET /v1/hivecomply/framework/map.

Live Endpoints

Production API reference

All endpoints are live at https://receipts.thehiveryiq.com. Health checks don't need a login. Audit and scan endpoints need a signed request header. The evidence bundles you get back from a scan are the same files the auditor reviews.

GET https://receipts.thehiveryiq.com/v1/aegis/health System health check; returns component status
POST https://receipts.thehiveryiq.com/v1/aegis/control/audit Audit a single control; body: {"control_id":"CC6.1"}
POST https://receipts.thehiveryiq.com/v1/aegis/scan Full-fleet scan of all 158 controls; returns signed bundle
GET https://receipts.thehiveryiq.com/v1/aegis/score Live composite score and per-framework breakdown
GET https://receipts.thehiveryiq.com/v1/aegis/control/{id} Retrieve latest signed envelope for a specific control
GET https://receipts.thehiveryiq.com/v1/aegis/matrix Full control-to-framework mapping matrix
GET https://receipts.thehiveryiq.com/v1/hivecomply/framework/map Cross-framework control map; shows the 2.32x leverage data
GET https://receipts.thehiveryiq.com/v1/hivecomply/control/status Per-control status across all mapped frameworks
POST https://receipts.thehiveryiq.com/v1/hivecomply/event/ingest Ingest an external compliance event from a GRC tool
POST https://receipts.thehiveryiq.com/v1/hivecomply/bundle/export Export auditor-ready evidence bundle as a signed archive
POST https://receipts.thehiveryiq.com/v1/hivecomply/bundle/verify Verify the integrity and signatures of an exported bundle
POST https://receipts.thehiveryiq.com/v1/purity/cert/issue Issue a Purity receipt anchoring an attestation event
GET https://receipts.thehiveryiq.com/v1/purity/health Purity receipt subsystem health check

Verify the live score yourself

shell
# Retrieve the live composite score and framework breakdown
curl -s https://receipts.thehiveryiq.com/v1/aegis/score

# Audit a single control (CC6.1: Logical access control)
curl -s -X POST https://receipts.thehiveryiq.com/v1/aegis/control/audit \
  -H 'Content-Type: application/json' \
  -d '{"control_id":"CC6.1"}'

# Retrieve the full control-to-framework matrix
curl -s https://receipts.thehiveryiq.com/v1/aegis/matrix

# Export a signed evidence bundle
curl -s -X POST https://receipts.thehiveryiq.com/v1/hivecomply/bundle/export

Beyond SOC 2 Addendum

What AEGIS adds that the frameworks don't measure

SOC 2 and ISO 27001 set the floor you have to clear. They spell out minimum control requirements for security, availability, confidentiality, and privacy. But they don't cover what happens if today's cryptography breaks, how you prove your randomness source is trustworthy, or how you prove data was actually deleted. AEGIS covers all four of those gaps on purpose.

Deletion Proofs

Cryptographic deletion proofs

When GDPR Article 17 calls for erasure, or a customer asks you to delete their data, AEGIS-Report issues a signed deletion envelope. It contains the SHA3-256 digest of the data marked for erasure, a timestamp, the Ed25519 and ML-DSA-65 signatures over the deletion record, and an on-chain anchor through AEGIS-Seal. The result is proof anyone can check that data was erased, not just a person's word that it was erased. SOC 2 and ISO 27001 require you to write down your deletion procedures. Neither one requires cryptographic proof that deletion actually happened.

Post-Quantum Signatures

Post-quantum signatures per control

Every control evidence envelope in AEGIS carries an ML-DSA-65 signature (FIPS 204, the government's post-quantum signature standard) on top of the classical Ed25519 signature. ML-DSA-65 comes from the CRYSTALS-Dilithium family, chosen by NIST during its post-quantum cryptography standardization process. That means if a powerful enough quantum computer ever breaks today's elliptic-curve cryptography, the evidence chain still holds up under the post-quantum signature. SOC 2 and ISO 27001 mention cryptographic controls but don't require post-quantum readiness. AEGIS gives it to you by default, not as an option you have to turn on.

On-Chain Anchored Receipts

On-chain anchored attestation receipts

AEGIS-Seal locks each evidence bundle onto the Base L2 blockchain. The record holds the bundle hash and the block's timestamp. That timestamp comes from the Base network itself. Nobody can change it, not even Hive Civilization. That gives auditors a way to check things on their own: they can look at the on-chain record directly through the public Base block explorer and confirm the evidence bundle existed at the time it claims, without having to trust Hive Civilization's servers or records. No SOC 2 or ISO 27001 requirement covers on-chain timestamping. It's something AEGIS does on its own.

Entropy Provenance

Swarm-validated entropy provenance

The keys that sign AEGIS evidence envelopes come from randomness gathered through MAPET-X, a 24-source physics-based randomness system. MAPET-X pulls randomness from the ground (gravity, magnetic fields, seismic activity), light and quantum sources (photon arrival, polarization), subatomic events (radioactive decay, cosmic muons), biological signals (heart-rate variance, EEG), space (cosmic background radiation, solar wind), and even soil (microbial activity). Before any of that randomness gets used to build a key, T-THRESH checks it across a spread-out group of MAPET-X nodes. That group check means no single broken sensor can poison the randomness pool. SOC 2 and ISO 27001 cover key management broadly. Neither one covers where your randomness physically comes from or how it gets validated by a group of nodes.

Honest disclosure: what these capabilities are not

These four capabilities don't replace a formal SOC 2 Type 1 or Type 2 opinion, an ISO 27001 certification, or any other third-party audit. Cryptographic deletion proofs, post-quantum signatures, on-chain anchors, and swarm-checked randomness are engineering pieces that strengthen the evidence chain. They don't replace the decision of an independent auditor or certification body. Hive Civilization is aiming for a SOC 2 Type 1 engagement in Q4 2026. Right now the state is self-attested. See the SOC 2 Self-Attested inventory and ISO 27001 Self-Attested inventory for the full control-by-control disclosure.


Honest Disclosure

Machine-checked vs. self-attested

32.9% of the 158 controls are machine-checked. AEGIS probes them, hashes them, and signs them with no person involved. The other 67.1% are self-attested. A person reviewed the control, wrote down how it's set up, and signed off on it. We won't tell you otherwise.

Machine-checked (32.9%)

AEGIS checks these controls on its own. The check runs against your live system, captures the result, computes a SHA3-256 digest, and signs the envelope with Ed25519 and ML-DSA-65. The score for these controls reflects what the system actually saw, not what a person reported. This group includes: logical access setups (CC6.x), encryption state of data at rest and in transit, certificate validity, endpoint detection agent status, and infrastructure configuration baselines. Machine-checked controls get rechecked on every scan cycle (460 ms for the full fleet).

Self-attested (67.1%)

These controls need decision, policy review, or process checks that can't be turned into a simple pass or fail test. Examples include: governance controls (CC1.x, commitment to integrity), risk assessment processes (CC3.x), vendor management reviews (ISO 27036), and AI-specific governance items (ISO 42001, EU AI Act). A person reviewed each self-attested control against what it requires, wrote down the real state honestly, and signed off on it. Self-attested controls get reviewed every quarter. The next review is planned for Q3 2026.

What we won't tell you

We won't tell you that 96.2 means everything is fully compliant. It means that across all 158 controls mapped to 12 frameworks, the weighted score as of the last scan is 96.2 out of 100. Some controls are still just planned. Some are partly done. The full control-by-control inventory is at thehiveryiq.com/security/soc2-self-attested/ and thehiveryiq.com/security/iso-27001-self-attested/. We publish the honest state because an audit that only shows you green lights isn't really an audit.


TACHYON

Post-quantum signing: the TACHYON layer

Every AEGIS evidence envelope gets signed by two algorithms side by side. The classical path uses Ed25519. The post-quantum path uses ML-DSA-65 from the TACHYON family. Both signatures have to check out for the evidence to count. That's the multi-assumption approach: it stays secure unless all three underlying assumptions fail at the same time.

T-DSA

Parallel digital signatures

ML-DSA-65 (CRYSTALS-Dilithium, FIPS 204) plus SLH-DSA (stateless hash-based) plus ECDSA all run side by side. All three have to produce valid signatures. The AEGIS envelope carries ML-DSA-65 plus Ed25519 by default. SLH-DSA is there for archival-grade evidence that needs hash-only security.

T-VAULT

Quantum-resistant key vault

Signing keys for AEGIS evidence envelopes live in T-VAULT, a post-quantum key vault running on STM32N6 hardware with AES-256 at rest. Keys are built from MAPET-X randomness after T-THRESH checks it across a group of nodes. Keys get rotated without any service downtime through T-AGILITY.

T-AGILITY

Algorithm-agility runtime

As NIST's post-quantum standards change over time, T-AGILITY lets us swap the signing algorithm without a redeploy. The AEGIS envelope format includes an algorithm ID field. Old envelopes still check out under the algorithm that made them. New envelopes use the current default. No evidence gets stranded when we switch algorithms.

The full TACHYON setup, ten products, three assumption families, MAPET-X randomness grouping, and the physics-based key chain, is written up at thehiveryiq.com/tachyon/. Here's the multi-assumption idea in plain terms: a TACHYON-signed artifact stays secure unless the lattice math, the hash-function math, and the physics-based randomness all fail at the same time. The odds of all three independent things failing together is the product of each one's own odds of failing.


Spheres of Truth

Attestation receipt pricing tiers

Every attestation event can be anchored as a Purity receipt through POST /v1/purity/cert/issue. Receipt tiers reflect how durable and how widely spread the on-chain anchor is. Higher tiers spread to more nodes and last longer.

Sphere Price per receipt Description
Whisper $0.0001 Minimal on-chain anchor; single-node confirmation; suitable for high-frequency low-value events
Echo $0.0008 Two-node propagation; standard compliance event logging
Cipher $0.0012 Three-node propagation with ML-DSA-65 counter-signature on the anchor transaction
Thunder $0.0024 Five-node propagation; recommended for GDPR deletion proof anchoring
Cathedral $0.0048 Full-quorum propagation; suitable for control evidence anchoring in regulated industries
Constellation $0.0096 Cross-chain propagation; long-retention archival anchor for audit artifacts
Nebula $0.0192 Extended propagation with MAPET-X entropy provenance attestation attached
Galaxy $0.0384 Full-swarm propagation; SOC 2 Type 2 grade anchoring with T-THRESH validated entropy
Genesis $0.0768 Maximum tier; all nodes, all chains, full MAPET-X entropy certificate, 7-year retention guarantee


See the live score

96.2 and counting. See every number behind it.

The compliance dashboard shows the live score, per-framework breakdown, per-control evidence state, and the honest disclosure of what is machine-verified versus self-attested. Every number is sourced from a signed AEGIS envelope.


THE HIVE FAMILY

CRE is one surface. Here's the family it belongs to.

Every Hive surface signs its own evidence with the same primitives: SHA3-256 canonical hashing, Ed25519 + ML-DSA-65 dual signatures, and a published Merkle Mountain Range root. The receipt is the audit evidence. That same envelope works the same way everywhere: every transaction, every framework, every surface.