Skip to content

chore: add Amp orb setup script - #64

Merged
rmems merged 1 commit into
mainfrom
chore/amp-orb-setup
Oct 24, 2026
Merged

rmems merged 1 commit into
mainfrom
chore/amp-orb-setup

Conversation

@rmems

@rmems rmems commented Oct 2, 2026 •

Copy link
Copy Markdown
Owner

User description

What

Adds .agents/setup, the script Amp runs on a fresh orb before it saves the snapshot (docs). It's an executable bash script with set -euo pipefail that runs from the repo root. It's idempotent: on a re-run, anything already installed is skipped. It installs only what this repo's CI and manifests need, and ends with a dependency fetch or prebuild, not a test run, so it stays well under the 20-minute setup limit.

What it installs

  • apt: build-essential, cmake (Debian 12: 3.25), shellcheck
  • CPU-only configure: cmake -S kernels -B build/kernels-cpu -DBKL_ENABLE_CUDA=OFF

Pins and notes

  • Pins: cmake_minimum_required(3.24) (Debian 12 has 3.25.1); CPU path -DBKL_ENABLE_CUDA=OFF and the python3 tools/check_*.py / test_*.py checks from ci-cpu.yml use only the orb's system python3.
  • GPU: everything with -DBKL_ENABLE_CUDA=ON (kernel builds, benches, compute-sanitizer) needs the self-hosted [self-hosted, Linux, X64, CUDA] runner. On an orb, only the CPU checks and docs/tools work.
  • Not installed: actionlint 1.7.12 (ci-lint.yml downloads it per run).

No GPU on orbs

Amp orbs are Debian 12 cloud VMs with no GPU. The GPU-gated parts of this repo can't run on an orb; the notes above give the CPU path.

Testing

  • bash -n and shellcheck 0.10.0: clean.
  • Not run end to end. Only five of the scripts in this batch were run in a Debian 12 chroot (Limen-Capital, SynapticDistill.jl, combine-for-AI, Dioscuri-Cloud, silicon-hdl). This one has only been syntax-checked and linted.
  • Not run on a real Amp orb.

Things to double-check

  • rustup, juliaup and uv install into ~/.cargo/bin, ~/.juliaup/bin and ~/.local/bin and add those to the login profile. The script exports them for its own run. In a non-login, non-interactive shell they weren't on PATH, so check that Amp's agent shell finds cargo/julia/uv.
  • apt-get uses Acquire::Retries=5, but a deb.debian.org 500 can still fail the setup. Amp doesn't save a snapshot from a failed setup, so a retry fixes it.
  • Editing this script later doesn't rebuild existing orb snapshots. Delete it with amp projects snapshots delete <project> to force a fresh setup, or wait for it to expire (Amp reuses a snapshot for up to 72 hours).

Propose-only: please review before merging.


Summary by cubic

Adds .agents/setup, the bash script Amp runs on a fresh orb before saving a snapshot (Debian 12, CPU only). It installs only the packages this repo's CI needs — build-essential, cmake, shellcheck — and ends with the CPU-only CMake configure (-DBKL_ENABLE_CUDA=OFF), which skips every kernel target so no nvcc is required.

  • Idempotent: packages already installed are skipped on re-runs.
  • GPU-gated targets can't run on orbs, so setup stays on the CPU path.
  • Only syntax-checked and shellcheck-linted so far, not run end to end.

Written for commit 95c194a. Summary will update on new commits.

View guided diff Turn on auto-fix


CodeAnt-AI Description

Prepare Amp orbs for CPU-based development

What Changed

  • Fresh Amp orbs install the compiler, CMake, and ShellCheck needed for this repository.
  • The setup configures the CPU-only kernel path without requiring a GPU.
  • Already-installed packages are skipped when setup runs again.

Impact

✅ Ready-to-use CPU development on Amp orbs
✅ CPU setup without GPU hardware
✅ Repeatable orb setup

💡 Usage Guide

Checking Your Pull Request

Every time you make a pull request, our system automatically looks through it. We check for security issues, mistakes in how you're setting up your infrastructure, and common code problems. We do this to make sure your changes are solid and won't cause any trouble later.

Talking to CodeAnt AI

Got a question or need a hand with something in your pull request? You can easily get in touch with CodeAnt AI right here. Just type the following in a comment on your pull request, and replace "Your question here" with whatever you want to ask:

@codeant-ai ask: Your question here

This lets you have a chat with CodeAnt AI about your pull request, making it easier to understand and improve your code.

Example

@codeant-ai ask: Can you suggest a safer alternative to storing this secret?

Preserve Org Learnings with CodeAnt

You can record team preferences so CodeAnt AI applies them in future reviews. Reply directly to the specific CodeAnt AI suggestion (in the same thread) and replace "Your feedback here" with your input:

@codeant-ai: Your feedback here

This helps CodeAnt AI learn and adapt to your team's coding style and standards.

Example

@codeant-ai: Do not flag unused imports.

Retrigger review

Ask CodeAnt AI to review the PR again, by typing:

@codeant-ai: review

Check Your Repository Health

To analyze the health of your code repository, visit our dashboard at https://app.codeant.ai. This tool helps you identify potential issues and areas for improvement in your codebase, ensuring your repository maintains high standards of code health.

@codeant-ai

codeant-ai Bot commented Oct 2, 2026 •

Copy link
Copy Markdown

🤖 CodeAnt AI — Review Status

Status Commit Started (UTC) Finished (UTC)
✅ Incremental review completed 95c194a Oct 22, 2026 · 02:22 02:23
✅ Reviewed your PR 50128b1 Oct 02, 2026 · 01:12 01:13

@codeant-ai

codeant-ai Bot commented Oct 2, 2026

Copy link
Copy Markdown

Thanks for using CodeAnt! 🎉

We're free for open-source projects. if you're enjoying it, help us grow by sharing.

Share on X ·
Reddit ·
LinkedIn

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 2, 2026 •

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-02T01:13:12.991489Z 50128b1 PR opened
🔒 Security Review ✅ Completed 2026-10-02T01:14:40.784784Z 50128b1 PR opened
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" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@coderabbitai

coderabbitai Bot commented Oct 2, 2026 •

Copy link
Copy Markdown
Contributor

Review in Change Stack →

Important

Review skipped

Review was skipped as selected files did not have any reviewable changes.

⚙️ Run configuration
  • Configuration used: Organization UI
  • Review profile: ASSERTIVE
  • Plan: Advanced
  • Run ID: 63ec6f37-9838-4d7d-a0bd-3a842bba2da7
📥 Commits

Reviewing files that changed from the base of the PR and between 50128b1 and 95c194a.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

  • 🔍 Trigger review

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info
⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Advanced

Run ID: d928faab-171e-42ca-88b3-e0f31258f30f

📥 Commits

Reviewing files that changed from the base of the PR and between 22cc4f3 and 50128b1.

📒 Files selected for processing (1)
  • .agents/setup

Included review availability: This review used your included allowance. Your plan provides up to 1 included review per hour; 0 remain after this review.

📜 Recent review details
⏰ Context from checks skipped due to timeout. (3)
  • GitHub Check: Blocks PR Review
  • GitHub Check: Host GPU probe + kernel smoke
  • GitHub Check: Codacy Static Code Analysis
🧰 Additional context used
🔍 Remote MCP Linear

Linear

  • The parent BKL epic assigns GPU CI and CUDA validation to this repository; its release acceptance requires PR, CPU, and conditional GPU checks to pass. (Source: Linear · get_issue)
  • Linear’s PR record contains no linked issues, so no PR-specific acceptance criteria were available. (Source: Linear · get_diff)
  • Codex, CodeAnt, and CodeRabbit reviews were still running for commit 50128b1; no completed findings were present in the retrieved threads. (Source: Linear · get_diff_threads)

Retrieval limitations

Some requested queries to Linear were not sent; their results are unavailable.

🔇 Additional comments (1)
.agents/setup (1)

1-43: LGTM!

📝 Summary

Summary by CodeRabbit

  • Chores
    • Added a repeatable Debian 12 setup script that installs required build tools and configures CPU-only builds.

Adds a repeatable Debian 12 setup script. It installs missing build and lint packages, configures the kernels project for CPU-only builds, and prints a completion message. The script does not run tests.

Changes: Debian setup

Layer / File(s) Summary
Execution setup and package installation
.agents/setup
The script enables strict Bash behavior, changes to the repository root, and uses sudo for privileged commands when needed. It installs build-essential, cmake, and shellcheck only when missing.
CPU-only kernel configuration
.agents/setup
The script configures kernels in build/kernels-cpu with BKL_ENABLE_CUDA=OFF and prints a completion message.

Priority: ⬇️ Low

Estimated code review effort: 2 (Simple) | ~10 minutes

Change: Other

Merge Risk: ⚪ Minimal · up to 50128

The setup flow is ready to merge; normal host prerequisites such as sudo access and apt repository availability remain environmental requirements.

Architecture risk: 🔵 Low · up to 50128

The changed surface does not map to a changed system, dependency edge, entrypoint, or external dependency.

Changed systems: None identified.

Architecture concerns
No architecture-level concerns identified.

Review details

Before / after behavior

  • observed — Modified behavior in .agents/setup: Adds a strict-mode Bash setup script that changes to the repository root and uses sudo for privileged commands when the current user is not root.
  • observed — Modified behavior in .agents/setup: Adds apt_install, which checks package installation status and, only when packages are missing, updates apt metadata and installs them noninteractively with retries. It is invoked for build-essential, cmake, and shellcheck.
  • observed — Modified behavior in .agents/setup: Adds a CPU-only CMake configure for kernels in build/kernels-cpu with BKL_ENABLE_CUDA=OFF, followed by a completion message; no test run is added.
🚥 Pre-merge checks | ✅ 5
✅ Passed checks (5 passed)
Check name Status Explanation
Title check ✅ Passed The title clearly and concisely identifies the main change: adding an Amp orb setup script.
Description check ✅ Passed The description explains the Amp orb setup script, its dependencies, CPU-only configuration, limitations, and testing status. It is directly related to the changeset.
Docstring Coverage ✅ Passed No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0…
Linked Issues check ✅ Passed Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check ✅ Passed Check skipped because no linked issues were found for this pull request.
✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Commit to this branch
  • Create a new PR
  • Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Autopilot is currently an internal CodeRabbit preview.

A rabbit checks the tools in place
Then readies kernels for a build
With CUDA off, the CPU waits
The setup script completes its task
And leaves the test run for another day

Comment @coderabbitai help to get the list of available commands.

@codeant-ai codeant-ai Bot added the size:M This PR changes 30-99 lines, ignoring generated files label Oct 2, 2026

@amazon-q-developer amazon-q-developer Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This PR adds a well-structured Amp orb setup script with proper bash safety practices (set -euo pipefail), idempotent package installation, and appropriate error handling. The script correctly handles sudo detection and package management for the CPU-only build path. No blocking issues identified.


You can now have the agent implement changes and create commits directly on your pull request's source branch. Simply comment with /q followed by your request in natural language to ask the agent to make changes.

@gitar-bot

gitar-bot Bot commented Oct 2, 2026 •

Copy link
Copy Markdown

Gitar is working

Verification

Rules No rules evaluated

Functional validation Not enabled

Set up

Auto-approval Not enabled

Set up

Gitar

@codacy-production

Copy link
Copy Markdown

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.

Adds .agents/setup for Amp orbs (Debian 12, no GPU). Installs only the toolchain pinned by CI and manifests and ends with a dependency fetch.
@rmems
rmems force-pushed the chore/amp-orb-setup branch from 50128b1 to 95c194a Compare October 22, 2026 02:22
@codeant-ai

codeant-ai Bot commented Oct 22, 2026

Copy link
Copy Markdown

CodeAnt PR Risk: Low Risk

  • The PR appears safe to merge; no actionable issues were identified in the review.

Assessed commit: 95c194abb621

@rmems rmems self-assigned this Oct 23, 2026
@rmems
rmems merged commit 0f14766 into main Oct 24, 2026
12 checks passed
@linear-code

linear-code Bot commented Oct 24, 2026

Copy link
Copy Markdown

RM-2335

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

size:M This PR changes 30-99 lines, ignoring generated files

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant