claude-code - 💡(How to fix) Fix [DOCS] Bedrock and Vertex setup wizard docs missing rerun details for config path, existing pins, and 1M options [1 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
anthropics/claude-code#49298Fetched 2026-04-17 08:45:13
View on GitHub
Comments
0
Participants
1
Timeline
4
Reactions
0
Author
Participants
Timeline (top)
labeled ×4
RAW_BUFFERClick to expand / collapse

Documentation Type

Missing documentation (feature not documented)

Documentation Location

https://code.claude.com/docs/en/amazon-bedrock

Section/Topic

The Bedrock and Vertex AI setup wizard documentation: the “Follow the wizard prompts” steps, the notes about re-running /setup-bedrock and /setup-vertex, and the related slash-command descriptions.

Current Documentation

https://code.claude.com/docs/en/amazon-bedrock currently says:

Choose how you authenticate to AWS: an AWS profile detected from your ~/.aws directory, a Bedrock API key, an access key and secret, or credentials already in your environment. The wizard picks up your region, verifies which Claude models your account can invoke, and lets you pin them. It saves the result to the env block of your user settings file, so you don't need to export environment variables yourself.

After you've signed in, run /setup-bedrock any time to reopen the wizard and change your credentials, region, or model pins.

https://code.claude.com/docs/en/google-vertex-ai currently says:

Choose how you authenticate to Google Cloud: Application Default Credentials from gcloud, a service account key file, or credentials already in your environment. The wizard detects your project and region, verifies which Claude models your project can invoke, and lets you pin them. It saves the result to the env block of your user settings file, so you don't need to export environment variables yourself.

After you've signed in, run /setup-vertex any time to reopen the wizard and change your credentials, project, region, or model pins.

Related docs cover the pieces separately, but not the wizard behavior itself:

CLAUDE_CONFIG_DIR — Override the configuration directory (default: ~/.claude). All settings, credentials, session history, and plugins are stored under this path.

If your account supports 1M context, the option appears in the model picker (/model) in the latest versions of Claude Code.

To enable extended context for a pinned model, append [1m] to the model ID in ANTHROPIC_DEFAULT_OPUS_MODEL or ANTHROPIC_DEFAULT_SONNET_MODEL.

What's Wrong or Missing?

A. The setup wizard pages do not explain what changes on re-run

The Bedrock and Vertex pages say you can reopen /setup-bedrock and /setup-vertex, but they do not document that re-running the wizard seeds model candidates from the pins that are already saved. That is useful rollout/admin behavior, especially when teams revisit pins after a model upgrade.

B. The setup wizard pages do not explain the actual settings file path when CLAUDE_CONFIG_DIR is set

The current wording only says the wizard saves to the “user settings file.” Separately, the env-var docs explain that CLAUDE_CONFIG_DIR relocates the config directory, but the setup guides never connect that to the wizard output. Users running multiple Claude Code profiles need to know the wizard writes to the real path under CLAUDE_CONFIG_DIR rather than always ~/.claude/settings.json.

C. The setup wizard pages do not document the new “with 1M context” choice for supported models

The model-config page documents /model 1M variants and the manual [1m] suffix for pinned models, but the Bedrock and Vertex setup guides do not say that the interactive provider setup flow can now offer a “with 1M context” option when a supported model is selected.

Changelog v2.1.111 describes all three of these as setup-wizard improvements, so the current setup documentation is incomplete for the user-facing behavior of /setup-bedrock and /setup-vertex.

Suggested Improvement

Add a short “What re-running the setup wizard updates” note to both provider pages and mirror the key points in https://code.claude.com/docs/en/commands:

  1. Clarify where the wizard writes pins and provider env vars:
  • default: ~/.claude/settings.json
  • when CLAUDE_CONFIG_DIR is set: <CLAUDE_CONFIG_DIR>/settings.json
  1. Document that re-running /setup-bedrock or /setup-vertex seeds the model candidates from existing saved pins instead of starting from a blank set.
  2. Document that supported models may show a “with 1M context” option in the wizard, and cross-reference that this maps to the same [1m] extended-context behavior described in the model configuration docs.

Even a short note or callout on each provider page would make the new v2.1.111 behavior discoverable without requiring users to infer it from separate CLAUDE_CONFIG_DIR and model-config pages.

Impact

Medium - Makes feature difficult to understand

Additional Context

Affected Pages:

PageContext
https://code.claude.com/docs/en/amazon-bedrockBedrock wizard setup steps and /setup-bedrock re-run note
https://code.claude.com/docs/en/google-vertex-aiVertex AI wizard setup steps and /setup-vertex re-run note
https://code.claude.com/docs/en/commandsSlash-command descriptions for /setup-bedrock and /setup-vertex
https://code.claude.com/docs/en/env-varsCross-reference for CLAUDE_CONFIG_DIR remapping of the config directory
https://code.claude.com/docs/en/model-configCross-reference for [1m] model pinning and 1M-context options

Total scope: 5 pages affected

Source: Changelog v2.1.111

Changelog entry:

Improved /setup-vertex and /setup-bedrock to show the actual settings.json path when CLAUDE_CONFIG_DIR is set, seed model candidates from existing pins on re-run, and offer a "with 1M context" option for supported models

extent analysis

TL;DR

Update the Bedrock and Vertex AI setup wizard documentation to include information on what changes when re-running the wizard, the actual settings file path when CLAUDE_CONFIG_DIR is set, and the new "with 1M context" choice for supported models.

Guidance

  • Add a note to both provider pages explaining what re-running the setup wizard updates, including the default and CLAUDE_CONFIG_DIR-set settings file paths.
  • Document that re-running /setup-bedrock or /setup-vertex seeds model candidates from existing saved pins.
  • Include information on the "with 1M context" option for supported models and cross-reference the model configuration docs.
  • Update the affected pages, including the Bedrock and Vertex AI setup pages, commands page, env-vars page, and model-config page.

Example

No code snippet is necessary for this documentation update.

Notes

The suggested improvements are based on the changelog entry for v2.1.111, which describes the new behavior of the setup wizards. The updates should make the feature more understandable and easier to use.

Recommendation

Apply the suggested improvements to the affected pages to ensure that users have accurate and complete information about the setup wizard behavior. This will help reduce confusion and make it easier for users to understand and use the feature.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING