
Argent Strategic Intelligence
ARGENT FORMILITARY & DEFENSE
Master Control System · by Nyedis
01 / ARGENT FOR MILITARY & DEFENSE
AI agents now act like staff
An AI agent is software that works for an employee: it reads, decides and acts using an AI model such as GPT, Claude or Gemini.
- Congress has ordered DeepSeek AI removed from Defense Department systems, and barred contractors from using it on Defense Department contract work.¹
- Under the arms export law, the State Department's civil penalty for each violation can reach the greater of $1,271,078 or twice the value of the deal.²
You used to secure people who use software. Now you also secure software that acts like a person.
02 / ARGENT FOR MILITARY & DEFENSE
Five ways it goes wrong
Your data-loss, web-proxy, identity and log tools were built for people and files, so they usually never see an agent's request.
- Controlled data leaves in a prompt. An engineer's coding agent sends a design note with an export marking to a model online. That can start a reporting clock: 72 hours from discovery.³
- Unapproved AI spreads. The model may be one Congress has barred. IBM found incidents involving unapproved ("shadow") AI rose to 43% of organizations from 20%, averaging $5.39 million each.⁴
- Agents reach too far. 92% of organizations with an AI-related breach lacked proper AI access controls.⁴
- Agents outlive their owners. The engineer leaves the program; the agent keeps reading program files, and no one answers for it.
- No proof. When an assessor asks "what did your agents do, and can you prove the record was never edited", the answer is a guess.
03 / ARGENT FOR MILITARY & DEFENSE
The one idea: a checkpoint in front of every agent
Argent Strategic Intelligence sits between your AI agents and the models and tools they use, and decides each request before it goes out.
- It installs inside your own environment, even a closed enclave, and is built to run with no internet: the console loads nothing from outside the install, and detection runs right there.
- A model you run yourself can be registered as running on your own infrastructure, so the enclave gets AI of its own.
- It governs the agents registered with it. It connects to 17 AI providers and to 65 agent tools and frameworks, including Claude Code, Codex, Cursor and GitHub Copilot, plus any tool with a base-URL or proxy setting.
04 / ARGENT FOR MILITARY & DEFENSE
How Argent answers each risk
| Risk | What Argent does |
|---|---|
| 1. Data leaves | Reads each request first and replaces Social Security, card and bank account numbers with <REDACTED> before the model sees them. A defense pack adds CUI banners and portion marks, ITAR and EAR export-control legends, CAGE codes and national stock numbers (NSNs). Set to Deny, the request is refused naming the kind of marking, never the text. The record names the kind and count, never the value. |
| 2. Unapproved AI | The approved-model list starts empty: a model nobody approved is refused. A rule can move a denied model's calls to an approved one. A live list shows every agent with its owner. An export lists every agent and its owner. |
| 3. Reaching too far | Tools are approved one by one, and a new or changed tool waits for an admin's decision. Document lookups are filtered to what the asking person's owner may see (add-on). The agent never holds the provider's key. |
| 4. Orphaned agents | Every agent has a named human owner. When the owner is deactivated, the agent is refused on its very next request, and the Org chart shows it. An admin can reassign it to a new owner. |
| 5. No proof | Every decision goes on a signed record with each row linked to the one before. Signed: any edit is detected. An assessor checks an evidence package offline on their own machine and sees "TENANT PACKAGE OK". Operators can sign in with a CAC or PIV smart card, and every card sign-in, and every refused one, is recorded. |
05 / ARGENT FOR MILITARY & DEFENSE
What regulators already expect
No one in defense is waiting for an AI law. For defense contractors, these are the pressures a program is judged against.
- Arms Export Control Act, 22 U.S.C. 2778(c): anyone who willfully violates it "shall upon conviction be fined for each violation not more than $1,000,000 or imprisoned not more than 20 years, or both."⁵
- DFARS 252.204-7012: the contractor implements NIST SP 800-171 on covered contractor information systems, and must "rapidly report" cyber incidents to DoD, which "means within 72 hours of discovery of any cyber incident."³
- CMMC, DFARS 252.204-7021: "Have and maintain for the duration of the contract a current CMMC status" at the required level, and "Only process, store, or transmit FCI or CUI on contractor information systems that have a CMMC status at the CMMC level required."⁶
- False Claims Act, 31 U.S.C. 3729: knowingly submitting a false claim for payment carries three times the government's damages, plus a civil penalty of $14,308 to $28,619 per claim for penalties assessed after July 3, 2025.⁷
06 / ARGENT FOR MILITARY & DEFENSE
How a defense team starts
Watch first. Nothing changes for staff until the program decides.
- Watch. New business units start in watch mode: Argent records what it would have done and forwards the request unchanged.
- Enforce, area by area. Data-loss checks (defense markings included), models, tools, owners and others switch independently.
- Inside your walls. It runs in your Kubernetes cluster (k3s and RKE2 included) or as a single-node bundle, or as an air-gapped bundle carried in on media and verified before import. In Argent's FIPS build, its own Go services run on a FIPS 140-3 validated module (CMVP #5247).
Argent Strategic Intelligence