Uv bun bump - #1074
Conversation
- Updated versions for several dependencies in package.json: - "@anthropic-ai/claude-code" from ^2.1.71 to ^2.1.74 - "@augmentcode/auggie" from ^0.18.1 to ^0.19.0 - "@ccusage/codex" and "@ccusage/mcp" from ^18.0.9 to ^18.0.10 - "@github/copilot" from ^0.0.423 to ^1.0.5 - "@google/gemini-cli" from ^0.32.1 to ^0.33.1 - "@googleworkspace/cli" from ^0.8.0 to ^0.13.2 - "@kaitranntt/ccs" from ^7.52.2 to ^7.53.0 - "@openai/codex" from ^0.112.0 to ^0.114.0 - "@sourcegraph/amp" from ^0.0.1770734955-g93e689 to ^0.0.1773389256-ga2bc61 - "@tobilu/qmd" from ^1.1.5 to ^2.0.1 - "@typescript/native-preview" from ^7.0.0-dev.20260308.1 to ^7.0.0-dev.20260313.1 - "agent-browser" from ^0.17.0 to ^0.19.0 - "agentcash" from ^0.8.1 to ^0.9.2 - "ccusage" from ^18.0.9 to ^18.0.10 - "cline" from ^2.6.1 to ^2.7.0 - "openclaw" from ^2026.3.7 to ^2026.3.12 - "oxfmt" from ^0.36.0 to ^0.40.0 - "oxlint" from ^1.51.0 to ^1.55.0 - "turbo" from ^2.8.14 to ^2.8.16 - "vite" from ^7.3.1 to ^8.0.0 - Added "acpx" dependency with version ^0.3.0. - Updated pyproject.toml to include "marker-pdf" dependency with version >=1.10.2. Entire-Checkpoint: ec97fc3467eb
Entire-Checkpoint: 7e9b6855fc9f
Entire-Checkpoint: 5a0e03d4090b
Entire-Checkpoint: b4ffd3ff6f36
Entire-Checkpoint: 1cc70fce1de0
Entire-Checkpoint: b185be8c0859
|
You do not have enough credits to review this pull request. Please purchase more credits to continue. |
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (2)
✅ Files skipped from review due to trivial changes (1)
📝 WalkthroughSummary by CodeRabbit
WalkthroughThis PR updates model registries and templates (introducing GPT 5.3, Gemini and Claude revisions, Minimax, GLM renames), adds template files and Fish function templates/tests, updates Fish functions to new model IDs, bumps dependencies, and registers new template mappings and a package URL in settings. Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 inconclusive)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
📝 Coding Plan
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
Summary of ChangesHello, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request focuses on updating and standardizing the configuration of various large language models and their associated tooling. It includes bumping dependency versions for both JavaScript and Python packages, refining model definitions, and introducing a templating system for model configurations and Fish shell functions to streamline future updates and ensure consistency across environments. Highlights
Changelog
Activity
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
Mesa DescriptionTL;DRRefreshes LLM model presets and bumps dependencies, adds templated configurations, and updates fish helpers to default to What changed?
Description generated by Mesa. Update settings |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request updates numerous dependencies and refactors several configuration files to use a new templating system based on scripts/llm-update.sh. The introduction of templates is a good step towards better configuration management. However, I've found a critical inconsistency where the generated configuration file config/pi/models.json is out of sync with the model versions defined in the root models.json. This should be fixed by running the update script and committing the result.
| "gpt": "gpt-5.4", | ||
| "gpt-codex": "gpt-5.3-codex", | ||
| "gemini-pro": "gemini-3-pro-preview", | ||
| "gemini-flash": "gemini-3-flash-preview", | ||
| "gemini-pro": "gemini-3.1-pro-preview", | ||
| "gemini-flash": "gemini-3.1-flash-preview", |
There was a problem hiding this comment.
There are inconsistencies between the model versions in this file and the generated configuration in config/pi/models.json. The new templating system is a great addition, but for it to be effective, the generated files must be in sync with the source of truth (models.json).
Specifically:
gptis updated togpt-5.4here, butconfig/pi/models.jsonstill has an entry forgpt-5.2. The templateconfig/pi/models.tpl.jsonuses__GPT__, which should resolve togpt-5.4.gemini-prois updated togemini-3.1-pro-previewhere, butconfig/pi/models.jsonusesgemini-3-pro-preview.gemini-flashis updated togemini-3.1-flash-previewhere, butconfig/pi/models.jsonusesgemini-3-flash-preview.
It appears the scripts/llm-update.sh script was not run to propagate these changes to config/pi/models.json. Please run the script and commit the updated files to ensure consistency.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
config/pi/models.json (1)
26-62:⚠️ Potential issue | 🟠 MajorGenerated model catalog is out of sync with
models.jsonsource-of-truth.Line 26 (
gpt-5.2) and Line 43/Line 60 (gemini-3-...) don’t match the current mappings in rootmodels.json(gpt-5.4,gemini-3.1-...). Since this file is template-generated, this should be regenerated or corrected to avoid stale ids.🛠️ Sync fix
- "id": "gpt-5.2", - "name": "GPT 5.2 [ChatGPT Pro]", + "id": "gpt-5.4", + "name": "GPT 5.4 [ChatGPT Pro]", ... - "id": "gemini-3-pro-preview", - "name": "Gemini 3 Pro", + "id": "gemini-3.1-pro-preview", + "name": "Gemini 3.1 Pro", ... - "id": "gemini-3-flash-preview", - "name": "Gemini 3 Flash", + "id": "gemini-3.1-flash-preview", + "name": "Gemini 3.1 Flash",🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@config/pi/models.json` around lines 26 - 62, The generated model entries in this file are stale: update the model IDs and names to match the source-of-truth mappings (replace "gpt-5.2" with the current "gpt-5.4" mapping and update "gemini-3-pro-preview"/"gemini-3-flash-preview" to the canonical "gemini-3.1-…" variants), or regenerate the template so the entire block matches root models.json; ensure you adjust related fields (name, reasoning, contextWindow, maxTokens, cost) for the corresponding symbols "gpt-5.2", "gemini-3-pro-preview", and "gemini-3-flash-preview" to the correct values from the authoritative models.json.
🧹 Nitpick comments (2)
config/pi/settings.json (1)
24-26: Consider pinning external package reference to a specific commit or tag.The external package URL lacks version pinning, which could lead to unexpected behavior if the upstream repository changes. For reproducibility and security, consider pinning to a specific commit hash or release tag:
"packages": [ - "https://github.com/davebcn87/pi-autoresearch" + "https://github.com/davebcn87/pi-autoresearch@v1.0.0" ],Alternatively, if tags are not available, use a commit hash.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@config/pi/settings.json` around lines 24 - 26, The packages array currently references an unpinned external repo URL ("https://github.com/davebcn87/pi-autoresearch"); update the entry under the "packages" key to pin to a specific release tag or commit hash (for example by using a GitHub tag ref or commit SHA in the URL or a git+ URL with @<tag_or_sha>) so the dependency is immutable and reproducible; locate the string "https://github.com/davebcn87/pi-autoresearch" and replace it with a URL that includes the chosen tag or commit (or an archive URL referencing the commit) and commit the change.config/pi/models.tpl.json (1)
168-168: Prefer templating the OpenRouter preset model id to avoid drift.Line 168 is hardcoded while the rest of this file is placeholder-driven.
♻️ Suggested change
- "id": "@preset/glm-4-7", + "id": "@preset/__GLM_NONDOT__",🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@config/pi/models.tpl.json` at line 168, The "id" field is hardcoded as "@preset/glm-4-7" causing drift; replace that literal with the same templating placeholder pattern used elsewhere in this template (e.g., the existing placeholder token for OpenRouter preset model) so the line becomes a template-driven value instead of "@preset/glm-4-7"; update the placeholder name to something descriptive like openrouter_preset_model_id (matching existing placeholder conventions used in this file) and ensure callers provide that value.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Outside diff comments:
In `@config/pi/models.json`:
- Around line 26-62: The generated model entries in this file are stale: update
the model IDs and names to match the source-of-truth mappings (replace "gpt-5.2"
with the current "gpt-5.4" mapping and update
"gemini-3-pro-preview"/"gemini-3-flash-preview" to the canonical "gemini-3.1-…"
variants), or regenerate the template so the entire block matches root
models.json; ensure you adjust related fields (name, reasoning, contextWindow,
maxTokens, cost) for the corresponding symbols "gpt-5.2",
"gemini-3-pro-preview", and "gemini-3-flash-preview" to the correct values from
the authoritative models.json.
---
Nitpick comments:
In `@config/pi/models.tpl.json`:
- Line 168: The "id" field is hardcoded as "@preset/glm-4-7" causing drift;
replace that literal with the same templating placeholder pattern used elsewhere
in this template (e.g., the existing placeholder token for OpenRouter preset
model) so the line becomes a template-driven value instead of "@preset/glm-4-7";
update the placeholder name to something descriptive like
openrouter_preset_model_id (matching existing placeholder conventions used in
this file) and ensure callers provide that value.
In `@config/pi/settings.json`:
- Around line 24-26: The packages array currently references an unpinned
external repo URL ("https://github.com/davebcn87/pi-autoresearch"); update the
entry under the "packages" key to pin to a specific release tag or commit hash
(for example by using a GitHub tag ref or commit SHA in the URL or a git+ URL
with @<tag_or_sha>) so the dependency is immutable and reproducible; locate the
string "https://github.com/davebcn87/pi-autoresearch" and replace it with a URL
that includes the chosen tag or commit (or an archive URL referencing the
commit) and commit the change.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 79a5fbc0-ab96-4668-b998-66cd4d4ef5ee
⛔ Files ignored due to path filters (1)
bun.lockis excluded by!**/*.lock
📒 Files selected for processing (11)
config/pi/models.jsonconfig/pi/models.tpl.jsonconfig/pi/settings.jsonhome-manager/programs/fish/functions/_coxe_function.fishhome-manager/programs/fish/functions/_coxe_function.tpl.fishhome-manager/programs/fish/functions/_coxeh_function.fishhome-manager/programs/fish/functions/_coxeh_function.tpl.fishmodels.jsonpackage.jsonpyproject.tomlscripts/llm-update.sh
There was a problem hiding this comment.
Pull request overview
Updates the repo’s “global tools” and LLM model configuration plumbing by bumping Bun/UV-tracked packages and extending the llm-update.sh templating system to cover Pi + fish Codex helper functions.
Changes:
- Bump Bun
package.jsontool dependencies and UVpyproject.tomltool list. - Update
models.json(GPT + Gemini mappings) and propagate model ID bumps in fish Codex helpers. - Add Pi model config templating (
config/pi/models.tpl.json) and include it inscripts/llm-update.sh.
Reviewed changes
Copilot reviewed 11 out of 12 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| scripts/llm-update.sh | Adds Pi + fish function templates to the model-substitution update pipeline. |
| pyproject.toml | Adds marker-pdf to UV-managed global tool dependencies. |
| package.json | Bumps/adjusts many Bun-installed CLI/tool dependencies and trustedDependencies. |
| bun.lock | Not updated in this PR (see comment). |
| models.json | Updates canonical model ID mappings (GPT + Gemini). |
| home-manager/programs/fish/functions/_coxeh_function.tpl.fish | New template for headless Codex helper using __GPT_CODEX__. |
| home-manager/programs/fish/functions/_coxeh_function.fish | Updates Codex model ID to gpt-5.3-codex. |
| home-manager/programs/fish/functions/_coxe_function.tpl.fish | New template for Codex helper using __GPT_CODEX__. |
| home-manager/programs/fish/functions/_coxe_function.fish | Updates Codex model ID to gpt-5.3-codex. |
| config/pi/settings.json | Adds external Pi package source configuration. |
| config/pi/models.tpl.json | Introduces Pi models template with model placeholders. |
| config/pi/models.json | Updates Pi models output, but currently inconsistent with template + models.json mappings (see comment). |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
You can also share your feedback on Copilot code review. Take the survey.
| @@ -40,60 +40,26 @@ | |||
| "maxTokens": 128000 | |||
| }, | |||
| { | |||
| "id": "gpt-5.1-codex-max", | |||
| "name": "GPT-5.1 Codex Max [ChatGPT Pro]", | |||
| "reasoning": true, | |||
| "input": [ | |||
| "text", | |||
| "image" | |||
| ], | |||
| "cost": { | |||
| "input": 1.75, | |||
| "output": 14, | |||
| "cacheRead": 0.175, | |||
| "cacheWrite": 0 | |||
| }, | |||
| "contextWindow": 400000, | |||
| "maxTokens": 128000 | |||
| }, | |||
| { | |||
| "id": "gpt-5.1-codex", | |||
| "name": "GPT-5.1 Codex [ChatGPT Pro]", | |||
| "reasoning": true, | |||
| "input": [ | |||
| "text", | |||
| "image" | |||
| ], | |||
| "cost": { | |||
| "input": 1.75, | |||
| "output": 14, | |||
| "cacheRead": 0.175, | |||
| "cacheWrite": 0 | |||
| }, | |||
| "contextWindow": 400000, | |||
| "maxTokens": 128000 | |||
| }, | |||
| { | |||
| "id": "gpt-5.1", | |||
| "name": "GPT-5.1 [ChatGPT Pro]", | |||
| "id": "gemini-3-pro-preview", | |||
| "name": "Gemini 3 Pro", | |||
| "keepRecentTokens": 20000 | ||
| }, | ||
| "packages": [ | ||
| "https://github.com/davebcn87/pi-autoresearch" |
| "packageManager": "bun@1.3.0", | ||
| "dependencies": { | ||
| "@anthropic-ai/claude-code": "^2.1.71", | ||
| "@augmentcode/auggie": "^0.18.1", | ||
| "@anthropic-ai/claude-code": "^2.1.74", | ||
| "@augmentcode/auggie": "^0.19.0", | ||
| "@biomejs/biome": "^2.4.6", | ||
| "@ccusage/codex": "^18.0.9", | ||
| "@ccusage/mcp": "^18.0.9", | ||
| "@ccusage/codex": "^18.0.10", | ||
| "@ccusage/mcp": "^18.0.10", | ||
| "@getgrit/cli": "^0.1.0-alpha.1743007075", | ||
| "@github/copilot": "^0.0.423", | ||
| "@google/gemini-cli": "^0.32.1", | ||
| "@github/copilot": "^1.0.5", | ||
| "@google/gemini-cli": "^0.33.1", | ||
| "@google/jules": "^0.1.42", | ||
| "@googleworkspace/cli": "^0.8.0", | ||
| "@kaitranntt/ccs": "^7.52.2", | ||
| "@googleworkspace/cli": "^0.13.2", | ||
| "@kaitranntt/ccs": "^7.53.0", | ||
| "@mariozechner/pi-coding-agent": "^0.57.1", | ||
| "@nanocollective/nanocoder": "^1.23.0", | ||
| "@openai/codex": "^0.112.0", | ||
| "@openai/codex": "^0.114.0", | ||
| "@pkieltyka/gitgenie": "^0.1.2", | ||
| "@pulumi/pulumi": "^3.225.1", | ||
| "@sourcegraph/amp": "^0.0.1770734955-g93e689", | ||
| "@sourcegraph/amp": "^0.0.1773389256-ga2bc61", | ||
| "@steipete/bird": "^0.8.0", | ||
| "@tobilu/qmd": "^1.1.5", | ||
| "@typescript/native-preview": "^7.0.0-dev.20260308.1", | ||
| "@tobilu/qmd": "^2.0.1", | ||
| "@typescript/native-preview": "^7.0.0-dev.20260313.1", | ||
| "@vibe-kit/grok-cli": "^0.0.34", | ||
| "agent-browser": "^0.17.0", | ||
| "agentcash": "^0.8.1", | ||
| "ccusage": "^18.0.9", | ||
| "acpx": "^0.3.0", | ||
| "agent-browser": "^0.19.0", | ||
| "agentcash": "^0.9.2", | ||
| "ccusage": "^18.0.10", | ||
| "clawdhub": "^0.3.0", | ||
| "cline": "^2.6.1", | ||
| "cline": "^2.7.0", | ||
| "mcporter": "^0.7.3", | ||
| "open-composer": "^0.8.23", | ||
| "openclaw": "^2026.3.7", | ||
| "oxfmt": "^0.36.0", | ||
| "oxlint": "^1.51.0", | ||
| "openclaw": "^2026.3.12", | ||
| "oxfmt": "^0.40.0", | ||
| "oxlint": "^1.55.0", | ||
| "ralph-tui": "^0.11.0", | ||
| "turbo": "^2.8.14", | ||
| "turbo": "^2.8.16", | ||
| "typescript": "^5.9.3", | ||
| "vite": "^7.3.1" | ||
| "vite": "^8.0.0" |
| ["config/ccs/glm.settings.tpl.json"]=config/ccs/glm.settings.template.json | ||
| ["config/codex/config.tpl.toml"]=config/codex/config.toml | ||
| ["config/cliproxyapi/config.tpl.yaml"]=config/cliproxyapi/config.template.yaml | ||
| ["config/pi/models.tpl.json"]=config/pi/models.json | ||
| ["home-manager/programs/fish/functions/_coxe_function.tpl.fish"]=home-manager/programs/fish/functions/_coxe_function.fish | ||
| ["home-manager/programs/fish/functions/_coxeh_function.tpl.fish"]=home-manager/programs/fish/functions/_coxeh_function.fish |
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
There was a problem hiding this comment.
4 issues found across 12 files
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="config/pi/models.json">
<violation number="1" location="config/pi/models.json:43">
P2: `config/pi/models.json` is out of sync with the source of truth in `models.json`. This should be `gemini-3.1-pro-preview` (matching `models.json`), not `gemini-3-pro-preview`. The same issue applies to `gemini-flash` below (`gemini-3-flash-preview` → `gemini-3.1-flash-preview`) and `gpt` (`gpt-5.2` → `gpt-5.4`). It looks like `scripts/llm-update.sh` wasn't re-run after updating `models.json` to propagate the template substitutions.</violation>
<violation number="2" location="config/pi/models.json:77">
P2: The new `claude-opus-4-6` entry still uses stale pricing and a 32K output limit instead of Opus 4.6's current specs.</violation>
</file>
<file name="config/pi/models.tpl.json">
<violation number="1" location="config/pi/models.tpl.json:26">
P2: Regenerate and commit `config/pi/models.json` with this template; Pi still loads the generated file directly, so these placeholder model IDs have no runtime effect yet.</violation>
</file>
<file name="config/pi/settings.json">
<violation number="1" location="config/pi/settings.json:25">
P2: Pin this git package to a specific tag or commit instead of tracking the repository's moving default branch.</violation>
</file>
Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review.
| "id": "claude-opus-4-6", | ||
| "name": "Claude Opus 4.6", | ||
| "reasoning": true, | ||
| "input": [ | ||
| "text", |
There was a problem hiding this comment.
P2: The new claude-opus-4-6 entry still uses stale pricing and a 32K output limit instead of Opus 4.6's current specs.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At config/pi/models.json, line 77:
<comment>The new `claude-opus-4-6` entry still uses stale pricing and a 32K output limit instead of Opus 4.6's current specs.</comment>
<file context>
@@ -108,8 +74,8 @@
{
- "id": "claude-opus-4-5-20251101",
- "name": "Claude Opus 4.5",
+ "id": "claude-opus-4-6",
+ "name": "Claude Opus 4.6",
"reasoning": true,
</file context>
| "maxTokens": 128000 | ||
| }, | ||
| { | ||
| "id": "__GPT__", |
There was a problem hiding this comment.
P2: Regenerate and commit config/pi/models.json with this template; Pi still loads the generated file directly, so these placeholder model IDs have no runtime effect yet.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At config/pi/models.tpl.json, line 26:
<comment>Regenerate and commit `config/pi/models.json` with this template; Pi still loads the generated file directly, so these placeholder model IDs have no runtime effect yet.</comment>
<file context>
@@ -0,0 +1,186 @@
+ "maxTokens": 128000
+ },
+ {
+ "id": "__GPT__",
+ "name": "__GPT_PRETTY__ [ChatGPT Pro]",
+ "reasoning": true,
</file context>
| "keepRecentTokens": 20000 | ||
| }, | ||
| "packages": [ | ||
| "https://github.com/davebcn87/pi-autoresearch" |
There was a problem hiding this comment.
P2: Pin this git package to a specific tag or commit instead of tracking the repository's moving default branch.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At config/pi/settings.json, line 25:
<comment>Pin this git package to a specific tag or commit instead of tracking the repository's moving default branch.</comment>
<file context>
@@ -21,6 +21,9 @@
"keepRecentTokens": 20000
},
+ "packages": [
+ "https://github.com/davebcn87/pi-autoresearch"
+ ],
"skills": [],
</file context>
| { | ||
| "id": "gpt-5.1", | ||
| "name": "GPT-5.1 [ChatGPT Pro]", | ||
| "id": "gemini-3-pro-preview", |
There was a problem hiding this comment.
P2: config/pi/models.json is out of sync with the source of truth in models.json. This should be gemini-3.1-pro-preview (matching models.json), not gemini-3-pro-preview. The same issue applies to gemini-flash below (gemini-3-flash-preview → gemini-3.1-flash-preview) and gpt (gpt-5.2 → gpt-5.4). It looks like scripts/llm-update.sh wasn't re-run after updating models.json to propagate the template substitutions.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At config/pi/models.json, line 43:
<comment>`config/pi/models.json` is out of sync with the source of truth in `models.json`. This should be `gemini-3.1-pro-preview` (matching `models.json`), not `gemini-3-pro-preview`. The same issue applies to `gemini-flash` below (`gemini-3-flash-preview` → `gemini-3.1-flash-preview`) and `gpt` (`gpt-5.2` → `gpt-5.4`). It looks like `scripts/llm-update.sh` wasn't re-run after updating `models.json` to propagate the template substitutions.</comment>
<file context>
@@ -40,60 +40,26 @@
- {
- "id": "gpt-5.1",
- "name": "GPT-5.1 [ChatGPT Pro]",
+ "id": "gemini-3-pro-preview",
+ "name": "Gemini 3 Pro",
"reasoning": true,
</file context>
| "id": "gemini-3-pro-preview", | |
| "id": "gemini-3.1-pro-preview", |
Entire-Checkpoint: f5b57385910c
Summary by cubic
Refreshes LLM model presets and bumps deps. Adds templated configs, updates fish helpers to default to
gpt-5.3-codex, adds tests, and a script hook to render templates.New Features
gpt→gpt-5.4,gpt-codex→gpt-5.3-codex, Gemini →gemini-3.1-*, Claude →*-4-6, addedminimax-m2.5, and cleaned up GLM naming.config/pi/models.tpl.jsonand fish templates (_coxe_function.tpl.fish,_coxeh_function.tpl.fish);scripts/llm-update.shnow renders these to targets.cxe,coxeh) now usegpt-5.3-codexand support prompt-friendly templates.https://github.com/davebcn87/pi-autoresearchinsettings.jsonpackages.bun.lock.Dependencies
@anthropic-ai/claude-code,@augmentcode/auggie,@ccusage/codex,@ccusage/mcp,@github/copilot(1.x),@google/gemini-cli,@googleworkspace/cli,@openai/codex,@sourcegraph/amp,@tobilu/qmd(2.x),@typescript/native-preview,cline,openclaw,oxfmt,oxlint,turbo,vite(8.0.0), and more.acpxand@pkieltyka/gitgenie.marker-pdf>=1.10.2.Written for commit 3101d69. Summary will update on new commits.