Continue CLI support provider Ask Sage

[alex-mcgraw-askSage]

Description

With a custom config.yaml file, a user can now specify models to use the Ask Sage provider (specifically, provider name is specified as 'askSage' in the provider parameter). Users can now use the CLI tool along with the VSCode extension, and can specify what tools/capabilities they would like the models to have within the CLI, as long as that model has the ability to use tools.

AI Code Review

• Team members only: AI review runs automatically when PR is opened or marked ready for review
• Team members can also trigger a review by commenting @continue-review

Checklist

• [✅] I've read the contributing guide
• [✅] The relevant docs, if any, have been updated or created - config.yaml file specification already in the Ask Sage documentation. (PR Update asksage.mdx #9757 )
• [✅] The relevant tests, if any, have been updated or created

Screen recording or screenshot

Tests

11 tests were created in "packages/openai-adapters/src/test/asksage-adapter.vitest.ts".

---

Continue Tasks: ▶️ 2 queued — View all

---

Summary by cubic

Add AskSage provider to the CLI so you can run AskSage models with tools via config.yaml. This aligns CLI behavior with the VS Code extension.

• New Features

  • Added AskSageApi adapter with chat completions and tool call support; streaming simulated via chunked responses. Streamed tool_calls now emit distinct indices.
  • Wired provider: "askSage" in index.ts and added AskSageConfigSchema.
  • Auth supports apiKey directly or email-based session token (env.email + env.userApiUrl) with token caching, 401 refresh, and cache reset on token-creation failure.
  • Explicit errors for unsupported APIs: embeddings, rerank, legacy completions, and FIM.
  • Added vitest coverage for requests, tools, streaming, and errors.

• Migration

  • Set provider: "askSage" and apiKey in config.yaml.
  • Optionally set env.email and env.userApiUrl to use a session token.
  • Pass tools in CLI only if the chosen model supports tools.

^{Written for commit b4d3844. Summary will update on new commits.}

[cubic-dev-ai]

2 issues found across 4 files

Prompt for AI agents (all issues)

Check if these issues are valid — if so, understand the root cause of each and fix them.


<file name="packages/openai-adapters/src/apis/AskSage.ts">

<violation number="1" location="packages/openai-adapters/src/apis/AskSage.ts:155">
P1: Token retrieval failures are cached for an hour, preventing retries after a transient error.</violation>

<violation number="2" location="packages/openai-adapters/src/apis/AskSage.ts:364">
P2: Streaming tool call deltas reuse index 0 for every call, collapsing multiple tool calls into one when merging streamed responses.</violation>
</file>

_{Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.}

[RomneyDa]

Addition of Ask Sage to open ai adapters.

@alex-mcgraw-askSage note since this package is a dependency of core you could deduplicate the AskSageResponse, AskSageRequestArgs, etc. types between the two by importing them from @continuedev/openai-adapters. This is non-blocking.

[sestinj]

🎉 This PR is included in version 1.8.0 🎉

The release is available on:

• npm package (@latest dist-tag)
• GitHub release

Your semantic-release bot 📦🚀