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

Tabularis Local PR Review

by TabularisDB in TabularisDB/tabularis

Reviews a Tabularis pull request locally, judging the diff against the repo's rule files, verifying every claim in its description against the code, and running its tests on the real branch.

Apache-2.0Auto-check passedDevelopment

Install Tabularis Local PR Review

skills CLI
$ npx skills add TabularisDB/tabularis --skill tabularis-pr-review -a claude-code

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

GitHub CLI
$ gh skill install TabularisDB/tabularis tabularis-pr-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/TabularisDB/tabularis.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/tabularis-pr-review .claude/skills/tabularis-pr-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
tabularis-pr-review
GitHub stars
5.1k
Token cost
~1.4k tokens
SKILL.md length
622 words
Files
2
Skills in repo
2
Repo updated
First seen
Licence
Apache-2.0

At a glance

Reviews a Tabularis pull request locally, judging the diff against the repo's rule files, verifying every claim in its description against the code, and running its tests on the real branch.

  • Works in 8 steps: Pull the PR → Load the rules you'll judge against → Verify the PR's claims — don't take them… → …
  • Reviewing a Tabularis pull request before merging it
  • SKILL.md covers Overview, Workflow, Gotchas and Common Mistakes
  • Calls gh, cargo and git

What it does

The stated core principle is not to trust the PR description: pulling the PR's metadata and diff is followed by reading the .rules markdown files for every changed file's domain, which name common offenders such as calling setState synchronously inside useEffect, missing exhaustive effect dependencies, use of any in TypeScript, and new user-facing strings missing their i18n keys.

Each factual claim in the PR description is then checked directly against the code, for example grepping for the i18n keys it claims to use or confirming a new Tauri command is registered in the invoke handler, rather than accepted at face value. An impact-analysis step runs a graph-impact tool upstream from each changed symbol and flags high or critical blast radius.

Checking out the actual PR branch is called mandatory before running any tests, because testing against the reviewer's own branch silently reports the new test cases as missing rather than failing. The deliverable is a review plus a copy-paste manual test plan with staged real data, not just opinions on the diff.

When your agent uses it

  • Reviewing a Tabularis pull request before merging it
  • Verifying that a PR's description actually matches what the code does
  • Preparing a manual test plan with real data for a reviewer to exercise a PR

Example prompts

  • “Review PR #42 and verify its claims against the code.”
  • “Check this PR out locally and run its tests on the actual branch.”
  • “lass uns PR 17 reviewen und einen Testplan vorbereiten.”

Requirements

  • The gh CLI
  • A local checkout of the Tabularis repository

Workflow steps

8 steps, taken from the step headings in SKILL.md.

  1. Pull the PR
  2. Load the rules you'll judge against
  3. Verify the PR's claims — don't take them on faith
  4. Impact analysis (per AGENTS.md)
  5. Check out the PR branch — this is mandatory before running tests
  6. Run the PR's tests on that branch
  7. Prepare REAL test data + a manual test plan
  8. Verdict

What it can do on your machine

Read from SKILL.md and the folder at commit e98d1fb. 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:

    • gh
    • cargo
    • git
    • pnpm
    • kubectl

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

  • Network

    No URLs in SKILL.md. Its commands use gh, git, pnpm and kubectl, 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

Tabularis Local PR Review loads about 1.4k tokens when it runs. Until then it costs about 99 tokens; SKILL.md has 622 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~99
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k

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 TabularisDB/tabularis at commit e98d1fb, republished under its Apache-2.0 licence (© TabularisDB). 622 words, ~1,360 tokens.

Download SKILL.mdSave it as .claude/skills/tabularis-pr-review/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
tabularis-pr-review
description
Use when reviewing a Tabularis pull request locally — checking code quality against the repo's own rules, verifying the PR's claims, running its tests on the correct branch, and preparing real test data plus a manual test plan for the human. Use when the user says 'review PR #N', 'lass uns PR reviewen', 'check this PR', or wants to test a PR with real data before merging.

Tabularis Local PR Review

Overview

A read-everything-then-verify review for Tabularis PRs. Goes beyond a diff skim: judges the diff against the repo's own .rules/, confirms the PR's stated claims are true, runs the PR's tests on the PR branch, and stages real data so the human can manually exercise the change. The deliverable is a review plus a copy-paste test plan, not just opinions.

Core principle: Don't trust the PR description — verify every claim against the code, then prove the behavior with real data.

Workflow

Work top to bottom. Each step feeds the next.

1. Pull the PR
bash
gh pr view <N> --json title,body,author,baseRefName,headRefName,state,additions,deletions,changedFiles,labels
gh pr diff <N> --name-only      # scope
gh pr diff <N>                  # full diff
2. Load the rules you'll judge against

Read .rules/*.md for every changed file's domain before forming opinions. The common offenders:

  • react.md #2 — setState called synchronously in useEffect (async-after-await is fine). Very common in modal/form PRs.
  • react.md #1 — exhaustive useEffect/useMemo/useCallback deps.
  • typescript.md — no any.
  • rust.md — pure helpers extracted + unit-tested; mod.rs stays orchestration-only; public APIs re-exported on refactor.
  • general.md — English-only comments; new user-facing strings need i18n keys in src/i18n/locales/*.
3. Verify the PR's claims — don't take them on faith

For each factual assertion in the description, prove it:

  • "Uses common.search / common.noResults i18n keys" → grep '"search"\|"noResults"' src/i18n/locales/en.json.
  • "Removing searchable={false} enables search" → read the component default (grep searchable src/components/ui/Select.tsx).
  • "Reads default_port from the manifest" → check the type (src/types/plugins.ts) — is it nullable? What's the fallback?
  • New Tauri command → confirm it's registered in src-tauri/src/lib.rs invoke_handler.
4. Impact analysis (per AGENTS.md)

For each non-trivial symbol the PR touches, run gitnexus_impact({target, direction: "upstream"}) and note blast radius / risk. Flag HIGH/CRITICAL.

5. Check out the PR branch — this is mandatory before running tests

Your local checkout is almost never the PR branch, so its tests silently don't exist (cargo test <name> → "0 filtered out", vitest count looks normal but is missing the new cases). Always:

bash
git branch --show-current        # confirm you are NOT on the PR branch
gh pr checkout <N>               # handles forks; working tree must be clean

Tell the user which branch they were on so they can return (git checkout <their-branch>).

6. Run the PR's tests on that branch
bash
pnpm vitest run <changed test files>          # e.g. tests/utils/k8s.test.ts
cd src-tauri && cargo test --lib <module>::<test_mod>   # full path; partial names filter to 0

Report real pass counts, not "looks fine."

Show full SKILL.md (298 more words)Show less
7. Prepare REAL test data + a manual test plan

This is the highest-value step and the reason to review locally. Decide what real inputs the feature consumes, stage them, and verify the change's core command/logic against them directly before handing off. Pattern:

  1. Identify the external input (a K8s cluster, a DB, a file, an API).
  2. Stand it up reproducibly — write a manifest/seed to /tmp/pr<N>-*.{yaml,sql} covering every code path: the happy path, the path that must NOT trigger, and the rejected/edge path.
  3. Run the change's actual command against it to prove the backend logic (e.g. run the exact kubectl ... -o jsonpath=... the PR added and confirm output).
  4. Hand the user a table: input → expected behavior, one row per test case, plus a cleanup command.

Example (K8s dialog PR): minikube + manifest with a single-port service (auto-prefill should fire), a multi-port service (must not fire), and a pod (skipped).

8. Verdict

Concise: what's solid, numbered non-blocker notes (rule + practical impact), blockers separately, and an approve / approve-with-nits / request-changes call.

Gotchas

SymptomCause
cargo test → "0 passed, N filtered out"Wrong branch (PR code absent) or partial test name — use the full module::test_mod path.
Tests "pass" but new cases missingReviewing on your own branch, not the PR branch. Do step 5.
Number(v) on a cleared input → 0Empty form field coerces to 0; watch for it silently disabling defaults/guards.
Hardcoded fallback survives a "remove hardcoding" PRGrep the magic value in the diff's neighborhood — a ?? 3306 often lingers.

Common Mistakes

  • Reviewing the diff without reading .rules/ → missing the project's own conventions.
  • Trusting the PR description's claims instead of grepping for proof.
  • Running tests on the wrong branch and reporting false green.
  • Handing the user "test it manually" with no data staged — defeats the point of a local review.

© TabularisDB, 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

SKILL.md and 1 other file in .claude/skills/tabularis-pr-review of TabularisDB/tabularis.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit e98d1fb

Compare with similar skills

Tabularis Local PR 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.

Tabularis Local PR Review compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Tabularis Local PR Review this skillTabularisDB/tabularis5.1k—~1.4kAutomated safety check: PassApache-2.0
Resolve PR Reviewshencangsheng/easydb_app590—~2.4kAutomated safety check: PassMIT
Code Reviewoaslananka/kicad-mcp-pro120—~3.9kAutomated safety check: PassMIT
PR Babysitteropeninterpreter/openinterpreter69k3 repos~4.2kAutomated safety check: PassApache-2.0
MAUI PR Performance Analysisdotnet/maui23k—~2.4kAutomated safety check: PassMIT
Verify Candidate PR Fixstickerdaniel/linkedin-mcp-server3.8k—~2kAutomated safety check: PassApache-2.0

Similar skills

  • 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

  • Code Review

    oaslananka/kicad-mcp-pro

    A skill your agent uses for GitHub Copilot pull request and code reviews in oaslananka/kicad-mcp-pro.

    120 GitHub stars~3.9k tokensUpdated today
    DevelopmentAuto-check passed

  • PR Babysitter

    openinterpreter/openinterpreter

    Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.

    69k GitHub starsUsed in 3 repos~4.2k tokens
    DevelopmentAuto-check passed

  • MAUI PR Performance Analysis

    dotnet/maui

    Official

    Interprets pinned managed benchmark evidence for a dotnet/maui pull request and writes a narrative for the performance review workflow, without running or publishing anything.

    23k GitHub stars~2.4k tokensUpdated today
    DevelopmentAuto-check passed

  • Verify Candidate PR Fix

    stickerdaniel/linkedin-mcp-server

    Reviews a candidate pull request's diff and tests against its linked issue, with an optional live comparison against a previously captured baseline, without editing, merging or pushing.

    3.8k GitHub stars~2k tokensUpdated today
    DevelopmentAuto-check passed

  • Compound Engineering Code Review

    EveryInc/compound-engineering-plugin

    Runs a staged pull request or diff review using selected reviewer personas, checking the change against its stated intent and project standards before producing findings.

    25k GitHub stars~2k tokensUpdated yesterday
    DevelopmentAuto-check passed

More from TabularisDB/tabularis

  • Tabularis Database Driver Plugin

    TabularisDB/tabularis

    Guides building or updating a Tabularis database driver plugin in Rust that speaks JSON-RPC over stdio and follows the repository's rules.

    5.1k GitHub stars~2.1k tokensUpdated today
    Auto-check passed

Works with

Categories

Questions about Tabularis Local PR Review

What does Tabularis Local PR Review do?

Reviews a Tabularis pull request locally, judging the diff against the repo's rule files, verifying every claim in its description against the code, and running its tests on the real branch. rules markdown files for every changed file's domain, which name common offenders such as calling setState synchronously inside useEffect, missing exhaustive effect dependencies, use of any in TypeScript, and new user-facing strings missing their i18n keys.

When should I use Tabularis Local PR Review?

Tabularis Local PR Review fits situations like: reviewing a Tabularis pull request before merging it; verifying that a PR's description actually matches what the code does; preparing a manual test plan with real data for a reviewer to exercise a PR.

How do I install Tabularis Local PR Review in Claude Code?

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

How do I install Tabularis Local PR Review in Codex?

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

Can I use Tabularis Local PR 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 TabularisDB/tabularis --skill tabularis-pr-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/tabularis-pr-review, .gemini/skills/tabularis-pr-review, .github/skills/tabularis-pr-review and .opencode/skills/tabularis-pr-review in your project.

What does Tabularis Local PR Review need to run?

Going by SKILL.md and its folder, Tabularis Local PR Review needs the command-line tools its instructions call (gh, cargo, git, pnpm and kubectl). Our summary lists: The gh CLI; A local checkout of the Tabularis repository.

Does Tabularis Local PR Review access the network?

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

Is Tabularis Local PR 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 Tabularis Local PR Review use?

Tabularis Local PR 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 Tabularis Local PR Review use?

About 1.4k tokens (SKILL.md is roughly 5.4k 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 Tabularis Local PR Review?

Skills that share tags, products or a category with Tabularis Local PR Review: Resolve PR Review (shencangsheng/easydb_app, 590 stars), Code Review (oaslananka/kicad-mcp-pro, 120 stars), PR Babysitter (openinterpreter/openinterpreter, 69k stars) and MAUI PR Performance Analysis (dotnet/maui, 23k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Tabularis Local PR Review?

TabularisDB (a GitHub organization) maintains it in TabularisDB/tabularis, which has 5,142 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on October 9, 2026.

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