New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Review

by timeplus-io in timeplus-io/proton

Review a Proton pull request or diff for correctness, streaming semantics, checkpointing, compatibility, and performance.

Apache-2.0Auto-check passedDevelopment

Install Review

skills CLI
$ npx skills add timeplus-io/proton --skill review -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install timeplus-io/proton review --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/timeplus-io/proton.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/review .claude/skills/review && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
review
GitHub stars
2.3k
Token cost
~1.1k tokens
SKILL.md length
473 words
Files
1
Skills in repo
7
Repo updated
First seen
Licence
Apache-2.0

At a glance

Review a Proton pull request or diff for correctness, streaming semantics, checkpointing, compatibility, and performance.

  • Works in 5 steps: Correctness and safety → Streaming semantics and state recovery → Compatibility and user-visible behavior → …
  • The user mentions reviewing code
  • SKILL.md covers Arguments, Gather the change first, Review priorities and Proton-specific checklist, plus 4 more sections
  • Calls git and gh

What it does

Review is an agent skill from timeplus-io/proton. Review a Proton pull request or diff for correctness, streaming semantics, checkpointing, compatibility, and performance. Make sure to use this skill whenever the user mentions reviewing code, checking changes, PR feedback, or wants an opinion on a diff, even if they just say 'look at this PR' or 'is this change safe'.

Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Development, covering Pull requests. It works with SQL. The repository describes itself as: The Fastest Unified Streaming SQL Engine in a Single C++ Binary. ⚡ Millisecond latency. 100+ GB/s throughput. Continuously compute real-time context from streams, logs, metrics… The licence is Apache-2.0.

When your agent uses it

  • The user mentions reviewing code
  • Checking changes
  • Wants an opinion on a diff
  • Even if they just say look at this PR

Example prompts

  • “look at this PR”
  • “is this change safe”
  • “/review”

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Correctness and safety
  2. Streaming semantics and state recovery
  3. Compatibility and user-visible behavior
  4. Performance and resource usage
  5. Maintainability

What it can do on your machine

Read from SKILL.md and the folder at commit 1c47e60. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • git
    • gh

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git and gh, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Review loads about 1.1k tokens when it runs. Until then it costs about 82 tokens; SKILL.md has 473 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~82
When it runs · the whole SKILL.md, loaded when a task matches
~1.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from timeplus-io/proton at commit 1c47e60, republished under its Apache-2.0 licence (© timeplus-io). 473 words, ~1,083 tokens.

Download SKILL.mdSave it as .claude/skills/review/SKILL.md (or your agent's skills folder).
name
review
description
Review a Proton pull request or diff for correctness, streaming semantics, checkpointing, compatibility, and performance. Make sure to use this skill whenever the user mentions reviewing code, checking changes, PR feedback, or wants an opinion on a diff, even if they just say 'look at this PR' or 'is this change safe'.

Review

Arguments

  • $0 (required): PR number, branch name, or diff spec (for example 1234, my-branch, HEAD~3..HEAD)

Gather the change first

If $0 is a PR number:

bash
gh pr view "$0" --json title,body,baseRefName,headRefName,url
gh pr diff "$0"

If $0 is a branch name:

bash
git diff origin/develop..."$0"
git log --oneline origin/develop.."$0"

If $0 is a diff spec:

bash
git diff "$0"
git log --oneline "$0"

Read surrounding file context when the diff alone is insufficient.

Review priorities

  1. Correctness and safety
  2. Streaming semantics and state recovery
  3. Compatibility and user-visible behavior
  4. Performance and resource usage
  5. Maintainability

Proton-specific checklist

  • Streaming mode semantics are preserved:
    • SELECT FROM stream remains continuous.
    • SELECT FROM table(stream) remains historical/batch.
  • Stateful processors and materialized views preserve hasState(), checkpoint(), and recover() invariants when touched.
  • _tp_time, _tp_delta, window columns, and EMIT behavior stay internally consistent.
  • Proton fences are used only in upstream-synced code and never in src/Storages/Stream/ or namespace DB::Streaming.
  • Cluster dependency boundaries remain acyclic; read surrounding module context when touched code crosses modules.
  • Parser, AST, serialization, metadata, and storage format changes include compatibility reasoning and tests.
  • Query-plan or processor changes do not introduce hidden blocking work in prepare() paths.
  • Join/window/aggregation changes do not create obvious unbounded state growth or watermark regressions.
  • User-visible SQL behavior changes include tests and docs updates when appropriate.

What to look for

Correctness
  • Wrong query results, dropped rows, duplicate rows, or broken changelog semantics
  • Incorrect watermark or window-close behavior
  • Missing cleanup on error paths
  • Lifetime bugs, invalid references, iterator invalidation, or ownership confusion
Concurrency and recovery
  • Shared mutable state without synchronization
  • Deadlock-prone lock ordering changes
  • Checkpoint/recover mismatches
  • State mutations hidden in code paths that should stay deterministic or replay-safe
Performance
  • Allocations or copies in hot streaming paths
  • New per-event work in prepare()
  • Unbounded maps/vectors keyed by user data
  • Regressions in joins, windows, aggregation, or WAL/historical handoff
Show full SKILL.md (195 more words)Show less
Maintainability
  • Logic duplicated across streaming and historical paths
  • New magic constants that should be settings
  • Heavy includes or non-trivial implementations pushed into widely included headers

Tests expectation

For behavior changes, look for the smallest set that proves safety and the issue-specific regression path when an issue fix is claimed:

  • Issue-specific regression test, or explicit justification if it is infeasible
  • Positive path test
  • Negative/error-path test
  • Recovery or restart test when state/checkpointing is involved
  • Streaming SQL regression test for user-visible semantics

Output format

Findings

List findings first, ordered by severity.

  • Blocker - correctness, corruption, crash, data loss, invalid recovery, major compatibility break
  • Major - realistic bug risk, missing important tests, strong performance regression
  • Minor - clarity or robustness issue that should be fixed but is not merge-blocking

Each finding should include:

  • file and line
  • concrete impact
  • why it is a real problem
  • smallest reasonable fix
Then include
  • Missing context
  • Test coverage assessment
  • Proton-specific compliance notes
  • Final verdict: Approve, Request changes, or Block

Review rules

  • Be strict on real bugs, not style noise.
  • Ignore pure formatting-only diffs.
  • Prefer no comment over a weak or speculative comment.
  • When no meaningful findings exist, state that explicitly and mention residual risk or missing evidence.

© timeplus-io, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/review of timeplus-io/proton.

Open the folder on GitHubat commit 1c47e60

Compare with similar skills

Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Review compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Review this skilltimeplus-io/proton2.3k—~1.1kAutomated safety check: PassApache-2.0
SeekDB Code Reviewoceanbase/seekdb3.1k—~2.1kAutomated safety check: PassApache-2.0
Scoping Change Relevancejaktestowac/awesome-copilot-for-testers116—~2.5kAutomated safety check: PassMIT
Review PRapache/shardingsphere21k—~6.4kAutomated safety check: PassApache-2.0
Mariadb Operator Commentmariadb-operator/mariadb-operator1k—~2.2kAutomated safety check: PassApache-2.0
PR Bot Reviewspgplex/pgconsole155—~2.1kAutomated safety check: PassApache-2.0

Similar skills

  • SeekDB Code Review

    oceanbase/seekdb

    Reviews seekdb pull requests and diffs for real defects in correctness, resources, concurrency, security and tests, reporting only Blocker or Major findings.

    3.1k GitHub stars~2.1k tokensUpdated today
    DevelopmentAuto-check passed

  • Scoping Change Relevance

    jaktestowac/awesome-copilot-for-testers

    Classifies a diff into file tags and hunk tags - new public export, new endpoint, modified auth, SQL string, migration, added dependency, touched prompt - then maps each tag to the quality practices…

    116 GitHub stars~2.5k tokensUpdated 1 mo ago
    DevelopmentAuto-check passed

  • Review PR

    apache/shardingsphere

    Review Apache ShardingSphere or user-authorized downstream pull requests and PR discussions from public or authorized repository evidence.

    21k GitHub stars~6.4k tokensUpdated today
    DevelopmentAuto-check passed

  • Mariadb Operator Comment

    mariadb-operator/mariadb-operator

    Post a comment (or a formal PR review) to a GitHub issue or pull request in mariadb-operator/mariadb-operator.

    1k GitHub stars~2.2k tokensUpdated today
    DevelopmentAuto-check passed

  • PR Bot Reviews

    pgplex/pgconsole

    Triage, verdict, and reply to automated PR review comments from GitHub Copilot and Greptile, then iterate until the bots have nothing left to say.

    155 GitHub stars~2.1k tokensUpdated 1 mo ago
    DevelopmentAuto-check passed

  • Resolve PR Review

    shencangsheng/easydb_app

    Resolve pull request code review comments end-to-end. An agent skill from shencangsheng/easydb_app.

    590 GitHub stars~2.4k tokensUpdated 1 mo ago
    DevelopmentAuto-check passed

More from timeplus-io/proton

  • Alloc Profile

    timeplus-io/proton

    Analyze jemalloc or async-profiler allocation profiles in collapsed stack format.

    2.3k GitHub stars~489 tokensUpdated 17 days ago
    Auto-check passed

  • Create Worktree

    timeplus-io/proton

    Create a Proton git worktree with local submodule reuse. An agent skill from timeplus-io/proton.

    2.3k GitHub stars~812 tokensUpdated 17 days ago
    Auto-check passed

  • Build And Verify

    timeplus-io/proton

    Build, compile, run server/client/cluster, execute unit/stateless/SQL tests, verify results, and troubleshoot build/test failures.

    2.3k GitHub stars~1.3k tokensUpdated 17 days ago
    Auto-check passed

  • SQL Usage

    timeplus-io/proton

    Timeplus streaming SQL covering stream types, EMIT policies, window functions, JOINs, materialized views, external streams, and UDFs.

    2.3k GitHub stars~1.7k tokensUpdated 17 days ago
    Auto-check passed

  • CI Diagnostics

    timeplus-io/proton

    Diagnose Proton CI failures and performance comparison results from GitHub checks and uploaded reports.

    2.3k GitHub stars~685 tokensUpdated 17 days ago
    Auto-check passed

  • Cpp Coding

    timeplus-io/proton

    Write or review Timeplus/Proton C++20 code covering naming conventions, Proton fences, clang-format, IProcessor patterns, and checkpointing.

    2.3k GitHub stars~582 tokensUpdated 17 days ago
    Auto-check passed

Works with

Categories

Questions about Review

What does Review do?

Review a Proton pull request or diff for correctness, streaming semantics, checkpointing, compatibility, and performance. Review is an agent skill from timeplus-io/proton. Review a Proton pull request or diff for correctness, streaming semantics, checkpointing, compatibility, and performance.

When should I use Review?

Review fits situations like: the user mentions reviewing code; checking changes; wants an opinion on a diff; even if they just say look at this PR.

How do I install Review in Claude Code?

Run `npx skills add timeplus-io/proton --skill review -a claude-code`. Or copy the skill folder (.claude/skills/review in timeplus-io/proton) into .claude/skills/review in your project. Claude Code loads it when a task matches its description.

How do I install Review in Codex?

Run `npx skills add timeplus-io/proton --skill review -a codex`. Or copy the skill folder (.claude/skills/review in timeplus-io/proton) into .agents/skills/review in your project. Codex loads it when a task matches its description.

Can I use Review in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add timeplus-io/proton --skill review -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/review, .gemini/skills/review, .github/skills/review and .opencode/skills/review in your project.

What does Review need to run?

Going by SKILL.md and its folder, Review needs the command-line tools its instructions call (git and gh).

Does Review access the network?

SKILL.md contains no URLs. Its commands use git and gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Review safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Review use?

Review is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Review use?

About 1.1k tokens (SKILL.md is roughly 4.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Review?

Skills that share tags, products or a category with Review: SeekDB Code Review (oceanbase/seekdb, 3.1k stars), Scoping Change Relevance (jaktestowac/awesome-copilot-for-testers, 116 stars), Review PR (apache/shardingsphere, 21k stars) and Mariadb Operator Comment (mariadb-operator/mariadb-operator, 1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Review?

timeplus-io (a GitHub organization) maintains it in timeplus-io/proton, which has 2,265 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on September 20, 2026.

Source: timeplus-io/proton on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.