starred/shannon-KeygraphHQ
1
0
Fork 0
mirror of https://github.com/KeygraphHQ/shannon.git synced 2026-04-25 17:45:53 +03:00
Code Issues 18 Projects Releases Packages Wiki Activity

[GH-ISSUE #69] Different Anthropic base url #22

New issue
Closed
opened 2026-02-27 07:20:02 +03:00 by kerem · 2 comments
kerem commented 2026-02-27 07:20:02 +03:00
Owner
Copy link

Originally created by @abouolia on GitHub (Feb 6, 2026).
Original GitHub issue: https://github.com/KeygraphHQ/shannon/issues/69

Is there an env variable to change Anthropic base url to run it with different model like Kimi K2.5 without using OpenRouter?

Originally created by @abouolia on GitHub (Feb 6, 2026). Original GitHub issue: https://github.com/KeygraphHQ/shannon/issues/69 Is there an env variable to change Anthropic base url to run it with different model like Kimi K2.5 without using OpenRouter?
kerem closed this issue 2026-02-27 07:20:02 +03:00
kerem commented 2026-02-27 07:20:03 +03:00
Author
Owner
Copy link

@ezl-keygraph commented on GitHub (Feb 9, 2026):

@abouolia The project uses Claude Code Router, changing Anthropic base url may not work as expected. Currently we don't have support for Kimi K2.5.

The only experimental models supported are

Provider Models
OpenAI gpt-5.2, gpt-5-mini
OpenRouter google/gemini-3-flash-preview

This feature is experimental and unsupported. Output quality depends heavily on the model. Shannon is built on top of the Anthropic Agent SDK and is optimized and primarily tested with Anthropic Claude models. Alternative providers may produce inconsistent results (including failing early phases like Recon) depending on the model and routing setup.

<!-- gh-comment-id:3869943510 --> @ezl-keygraph commented on GitHub (Feb 9, 2026): @abouolia The project uses Claude Code Router, changing Anthropic base url may not work as expected. Currently we don't have support for Kimi K2.5. The only experimental models supported are Provider | Models -- | -- OpenAI | gpt-5.2, gpt-5-mini OpenRouter | google/gemini-3-flash-preview This feature is experimental and unsupported. Output quality depends heavily on the model. Shannon is built on top of the Anthropic Agent SDK and is optimized and primarily tested with Anthropic Claude models. Alternative providers may produce inconsistent results (including failing early phases like Recon) depending on the model and routing setup.
kerem commented 2026-02-27 07:20:03 +03:00
Author
Owner
Copy link

@Yash-xoxo commented on GitHub (Feb 9, 2026):

Hey @abouolia,

While PR #83 adds ANTHROPIC_BASE_URL support, there's an important caveat here.

As @ezl-keygraph mentioned, Shannon uses Claude Code Router under the hood, which is tightly coupled to Anthropic's API structure. So just changing the base URL might not work with Kimi K2.5 because:

  1. Different models have different API contracts (request/response formats)
  2. Shannon's prompts and agent workflows are optimized specifically for Claude's behavior
  3. The Claude Code Router expects Anthropic-specific endpoints and responses

What Actually Works

Currently, only these experimental models are supported (via OpenRouter):

  • OpenAI: gpt-5.2, gpt-5-mini
  • Google: gemini-3-flash-preview

To use those, you'd go through OpenRouter rather than changing the base URL directly.

If You Still Want to Try

You could technically set ANTHROPIC_BASE_URL to point to a Kimi-compatible endpoint, but expect issues:

  • Recon phase might fail early
  • Tool use/function calling format differences
  • Unpredictable results since Shannon isn't tested against it

If you're okay with experimental and potentially broken behavior, give it a shot - but the officially supported path is sticking with Claude models or using the explicitly supported alternatives through OpenRouter.

Would Kimi K2.5 support through OpenRouter work for your use case instead?

<!-- gh-comment-id:3872122021 --> @Yash-xoxo commented on GitHub (Feb 9, 2026): Hey @abouolia, While PR #83 adds `ANTHROPIC_BASE_URL` support, there's an important caveat here. As @ezl-keygraph mentioned, Shannon uses Claude Code Router under the hood, which is tightly coupled to Anthropic's API structure. So just changing the base URL might not work with Kimi K2.5 because: 1. Different models have different API contracts (request/response formats) 2. Shannon's prompts and agent workflows are optimized specifically for Claude's behavior 3. The Claude Code Router expects Anthropic-specific endpoints and responses ## What Actually Works Currently, only these experimental models are supported (via OpenRouter): - OpenAI: `gpt-5.2`, `gpt-5-mini` - Google: `gemini-3-flash-preview` To use those, you'd go through OpenRouter rather than changing the base URL directly. ## If You Still Want to Try You could technically set `ANTHROPIC_BASE_URL` to point to a Kimi-compatible endpoint, but expect issues: - Recon phase might fail early - Tool use/function calling format differences - Unpredictable results since Shannon isn't tested against it If you're okay with experimental and potentially broken behavior, give it a shot - but the officially supported path is sticking with Claude models or using the explicitly supported alternatives through OpenRouter. Would Kimi K2.5 support through OpenRouter work for your use case instead?
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
feat/npx-integration
Readme-Update
feat/telemetry
ctf-mode
v1.1.0
v1.0.0
Labels
Clear labels   
pull-request

Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
starred/shannon-KeygraphHQ#22
No description provided.