Integrations: AI model providers

Argent with Amazon Bedrock

Put Argent in front of Bedrock runtime. Argent reads the model from the URL, applies your model rules and swaps, redacts the request, inspects the streamed reply, enforces budgets and signs each call with the AWS credential it holds.

Amazon Bedrock logo
01 / INTEGRATIONS: AI MODEL PROVIDERS

What Argent does with Amazon Bedrock

Argent governs the four Bedrock runtime calls: Converse, ConverseStream, InvokeModel and InvokeModelWithResponseStream.

  • Model read from the URL: Bedrock names the model in the path, so Argent reads it there and applies the approved list.
  • Callers hold no AWS key: the gateway signs each call with SigV4, using the one IAM credential held in the credential broker.
  • Streams inspected, not altered: Argent decodes Bedrock's event-stream replies to inspect and meter them without re-encoding the provider's bytes.
  • Model Swap and classes: for Converse and ConverseStream callers, and for InvokeModel callers that send the Anthropic Messages format, swapped to another Anthropic model.
  • Budgets and record: every call priced, held against a budget and signed.
02 / INTEGRATIONS: AI MODEL PROVIDERS

How it connects

Three ways to point a workload at Argent, all tested. The caller holds an Argent key owned by a named person, never an AWS key.

  1. Two environment variables for any SDK that honours them, boto3 included: AWS_ENDPOINT_URL_BEDROCK_RUNTIME set to the Argent route and AWS_BEARER_TOKEN_BEDROCK set to the Argent agent key. Argent strips the key and signs the upstream call.
  2. Plain HTTPS with no AWS SDK: the Argent key as a Bearer token to the same route.
  3. Forward proxy for tools that only honour HTTPS_PROXY, or code that already signs SigV4 and cannot change: Argent replaces the signature and re-signs.

Or no Argent key at all. A workload can prove who it is with a signed AWS STS request from its own role, bound to an agent that already has a human owner.

The IAM credential sits in your vault, such as AWS Secrets Manager. An AWS service control policy that lets only Argent's gateway role call Bedrock, Bedrock Mantle included, makes a call that skips Argent fail. See AWS coverage.

03 / INTEGRATIONS: AI MODEL PROVIDERS

What you configure

Everything below is set once in the Argent console and applies to every agent that calls Amazon Bedrock through the gateway.

SettingWhereWhat it does
Approved modelsPolicy > Model and endpoint accessDefault deny: a call is allowed only when the model is on the approved list. The list is platform AND tenant AND agent, and each layer can only narrow the one above.
Access rulesSame pageAllow or deny by provider, route type, model, hosted tools, function tools and output type, for the tenant, a department, a person or one agent. The narrowest rule wins, and a 7-day impact preview shows what a draft would change.
Model SwapModel Swap pickerRuns an approved model in place of the one the agent asked for, including when a budget runs low, with no change to agent code.
Model classesPolicy > Model classesA name such as fast stands for an ordered list of approved models, with failover on connection errors, 5xx, 429, timeouts and retired models.
Sensitive dataDLPPer entity type: redact and record, record only, or block the request.
Budgets and capsBudgetsBudgets for the company, org, department, person and agent. Each request is priced and held against the budget before it leaves, and an emergency cap stops a runaway agent now.
ModePer controlOff, watch or enforce. Watch records what would have been refused and changes nothing on the wire.

Start in watch, read what would have been refused, then switch on enforce. See Watch Before You Enforce.

04 / INTEGRATIONS: AI MODEL PROVIDERS

One request, step by step

What happens between your agent and Bedrock, in order.

  1. The agent sends its request to its Argent route, carrying its own Argent key and nothing from Bedrock.
  2. Who is calling. Argent identifies the agent from its key and checks it may use this endpoint. The endpoint comes from the route the gateway matched, never from a header the caller sets.
  3. Which model. Argent reads the model from the URL path and checks it against the approved list and access rules before anything leaves.
  4. Swap or class. If a Model Swap rule or a model class applies, an approved model runs instead, and Argent translates the request and reply between provider formats where needed.
  5. Sensitive data. The request is scanned by a reader that knows the Converse and InvokeModel formats, and secrets and personal data are redacted, recorded or blocked under your policy. If the scan cannot run, the request is refused, never passed as clean.
  6. Budget. The request is priced and the amount is held against the agent's budget, so parallel calls cannot overshoot it.
  7. Signature. The gateway signs the call with SigV4 using the brokered AWS credential.
  8. Reply and record. Token usage is read from the reply, the cost is settled, and the signed record is written.
05 / INTEGRATIONS: AI MODEL PROVIDERS

What the record shows

Every Bedrock call leaves one signed record, allowed or refused.

  • One row per call: the agent and the person who owns it, the tenant, the model, the decision and the policy that made it, latency, and the cost, token and cache figures.
  • Tamper-evident: each row is hash-chained to the one before and signed with an ed25519 key held in Vault Transit, so an edited or missing row shows.
  • Real cost: cost comes from the usage block in Bedrock's own reply, priced from your price book in USD per million tokens.
  • No unpriced spend: an agent under a budget that calls a model with no price is refused, so a new model cannot slip past a cap.

Conversations can be opened and replayed from the record. See Session Replay and One-Click Audit Evidence.

06 / INTEGRATIONS: AI MODEL PROVIDERS

Related capabilities and goals

More AI model providers

Every provider below is governed through Argent.

  • OpenAI
  • Anthropic
  • Google Gemini
  • Azure OpenAI and Foundry
  • Mistral
  • DeepSeek
  • OpenRouter
  • GitHub Copilot
  • Kiro
  • Ollama
  • vLLM
  • NVIDIA NIM
  • LM Studio
  • Hugging Face TGI