Repository navigation
fix: default Haiku 5.5 to its measured context window - #8
LaurensBosscher wants to merge 3 commits into
Conversation
Deep @oh-my-pi/* imports fail under OMP 18.2.5 extension loading because host onLoad only rewrites legacy scopes onto in-process shims. Switch to @earendil-works root imports (proven on this host) and register claude-opus-5 so existing modelRoles/fallbackChains keep resolving.
…, Fable 5.1, Mythos) + max effort Derive the picker from the host's anthropic catalog instead of a hardcoded id list, keeping the 1M/200K variant entries. Measured [1m] entitlements stay in tables; unmeasured 1M-capable models default to 200K with a -1m alternate. Map thinking levels onto each model's catalog effort ladder (adds max). Fix unsuffixed entries throwing when their preferred window has no runtime.
|
Important
This repository does not receive automatic reviews because it has fewer than 10 stars.
 |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: dd1493b51e
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| import { buildSessionContext } from "@oh-my-pi/pi-coding-agent/session/session-context"; | ||
| import type { CompactionEntry } from "@oh-my-pi/pi-coding-agent/session/session-entries"; | ||
| import { compact } from "@oh-my-pi/pi-agent-core/compaction"; | ||
| import { StringEnum, type AssistantMessage, type AssistantMessageEventStream, type Context, type Model, type SimpleStreamOptions, type Tool, type Usage } from "@earendil-works/pi-ai"; |
There was a problem hiding this comment.
These value imports are supplied by the OMP 18 host's legacy-scope rewrite, but neither package.json nor bun.lock installs an @earendil-works/* package, while the published peer contract still accepts @oh-my-pi/* >=16.3.6 and the README declares no newer OMP minimum. On a supported host that predates that rewrite, module evaluation therefore stops here before the provider can register; either retain an import path for those hosts or raise and document the minimum host version.
Useful? React with 👍 / 👎.
| const efforts = model.thinking?.efforts; | ||
| if (!efforts?.length) return wanted === "xhigh" ? "max" : wanted as ClaudeCodeEffort; | ||
| for (let i = rank; i >= 0; i--) { | ||
| if (efforts.includes(CLAUDE_CODE_EFFORTS[i])) return CLAUDE_CODE_EFFORTS[i]; |
There was a problem hiding this comment.
When AskClaude is invoked with Sonnet 5 or Sonnet 4.6 and thinking: "max", their catalog ladder can omit max (the new Sonnet 4.6 fixture contains only low/medium/high), so this loop silently returns high. These models are explicitly special-cased above because Claude Code reports their top effort as max, and xhigh is correctly mapped to it, so the newly accepted explicit max option should also remain max rather than being weakened.
Useful? React with 👍 / 👎.
Changes
Stacked after #7: this branch includes earlier commits until their PRs merge.
Verification