Skip to content

bedrock iam/docs - #1165

Closed
whysumedh wants to merge 4 commits into
mainfrom
bedrock-iam/docs
Closed

whysumedh wants to merge 4 commits into
mainfrom
bedrock-iam/docs

Conversation

@whysumedh

Copy link
Copy Markdown
Member
  • docs: add auto_caching caveats and 1h TTL via default_params
  • chore: docs update
  • chore: old bedrock models warn
  • chore: update bedrock IAM permissions in docs

roh26it and others added 4 commits November 5, 2026 10:35
Document the cost trade-off and Anthropic 400 cases for auto_caching,
how to get a 1-hour TTL with default_params, and add auto_caching to
the config target-fields table. Applied to both the root and aigw trees.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Copilot AI balanced review requested due to automatic review settings November 11, 2026 01:13
@mintlify

mintlify Bot commented Nov 11, 2026 •

Copy link
Copy Markdown
Contributor

Preview deployment for your docs. Learn more about Mintlify Previews.

Project Status Preview Updated
portkey-docs 🟢 Ready View Preview Nov 11, 2026, 1:14 AM

💡 Tip: Enable Automations to automatically generate PRs for you.

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟡 Changes recommended

The assumed-role guides recommend resource restrictions that break Bedrock listing actions, and the prompt-caching matrix omits documented Bedrock Mantle support.

2 open issues
What changed in this PR

This PR updates Bedrock IAM setup and Anthropic prompt-caching guidance across Portkey’s current, legacy, and Prisma AIRS documentation.

Changes:

  • Adds bedrock:CountTokens to IAM examples and documents optional feature permissions.
  • Expands automatic-caching guidance with provider support, one-hour TTL configuration, and client-marker caveats.
  • Adds caching cross-links and documents auto_caching in configuration references.
File Description
virtual_key_old/​product/​enterprise-offering/​cloud-marketplace/​aws.mdx Adds CountTokens to the legacy marketplace policy.
virtual_key_old/​product/​ai-gateway/​virtual-keys/​bedrock-amazon-assumed-role.mdx Expands legacy assumed-role permissions guidance.
self-hosting/​hybrid-deployments/​aws/​marketplace.mdx Adds CountTokens to the marketplace policy.
self-hosting/​hybrid-deployments/​aws/​eks.mdx Adds CountTokens to the EKS policy.
product/​model-catalog/​connect-bedrock-with-amazon-assumed-role.mdx Documents core and optional Bedrock permissions.
product/​enterprise-offering/​cloud-marketplace/​aws.mdx Adds CountTokens to the enterprise marketplace policy.
product/​ai-gateway/​virtual-keys/​bedrock-amazon-assumed-role.mdx Expands virtual-key assumed-role permissions guidance.
product/​ai-gateway/​configs.mdx Links caching TTL and client-marker guidance.
integrations/​llms/​bedrock/​count-tokens.mdx Documents the required CountTokens permission.
integrations/​llms/​anthropic/​prompt-caching.mdx Expands caching support, TTL, and compatibility guidance.
integrations/​libraries/​claude-code-bedrock.mdx Adds Bedrock IAM prerequisites for Claude Code.
api-reference/​inference-api/​config-object.mdx Adds the auto_caching configuration field.
aigw/​self-hosting/​hybrid-deployments/​aws/​eks.mdx Adds CountTokens to the AIRS EKS policy.
aigw/​product/​model-catalog/​connect-bedrock-with-amazon-assumed-role.mdx Expands AIRS model-catalog IAM guidance.
aigw/​product/​ai-gateway/​virtual-keys/​bedrock-amazon-assumed-role.mdx Expands AIRS virtual-key IAM guidance.
aigw/​product/​ai-gateway/​configs.mdx Adds AIRS caching guidance links.
aigw/​integrations/​llms/​anthropic/​prompt-caching.mdx Expands AIRS caching support and configuration guidance.
aigw/​integrations/​libraries/​claude-code-bedrock.mdx Adds IAM prerequisites to the AIRS Claude Code guide.
aigw/​api-reference/​inference-api/​config-object.mdx Documents auto_caching in the AIRS reference.

Review effort: Balanced

| Claude Platform on AWS | Yes | Yes | Yes |
| Google Vertex AI | Yes | Yes | Yes |
| Azure AI Foundry | Yes | Yes <sup>1</sup> | Yes <sup>1</sup> |
| Amazon Bedrock | Yes (InvokeModel) <sup>2</sup> | No (Converse) | No |
}
```

Replace `<bucket>`, `<account_id>` and `<bedrock_service_role>` with your own values, and narrow each `Resource` to the model, inference profile and job ARNs you actually use.

This branch was successfully deployed

1 active deployment
staging — e31435c4 Deployed Nov 11, 2026 by mintlify[bot]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants