Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Orviqa Lite: Is your repo ready to promote?

    2

    Before you post your project anywhere, the README decides whether visitors stay.

    Free

    Security scanned
    Orviqa Lite: Is your repo ready to promote?

    Orviqa Lite: Is your repo ready to promote?

    Example session with this skill installed

    Is this repo ready to promote?

    • Read your context and instructions
    • Compiled the orviqa lite

    Illustrative example for a fictional project, acme/tidyjson, a small CLI that formats JSON files. The project, its files and its line numbers are made up to show the shape of the output.

    Ready after 2 fixes. Nothing critical is missing, but the quick start installs a package name that differs from the one in pyproject.toml, and the repository has no topics for GitHub search to match.

    Step 1: Health check

    In good shape. 1 important gap left, then it is down to refinements.

    • Missing (IMPORTANT): GitHub topics. Add 5 to 10 topics using terms your audience would search: language, problem domain, ecosystem.

    Step 2: README audit

    What it is: tidyjson is a Python CLI that rewrites JSON files with stable key order and indentation, and exits non-zero on invalid JSON so it can run in pre-commit.

    First impression

    The first 20 lines hold the name, a one-line description, two badges and a GIF of the tool reformatting a file. A visitor knows what it does within the first three lines.

    [IMPORTANT] INSTALLATION: the install command names a different package

    • What is wrong: line 14 reads pip install tidy-json, but pyproject.toml line 2 declares name = "tidyjson".

    Why it matters here: the first command a visitor copies installs nothing, or an unrelated package, before they have seen the tool run.

    • Fix: change line 14 to pip install tidyjson.

    [OPTIONAL] CALL_TO_ACTION: no next step after the example

    What is wrong: the README ends at the usage example on line 41, with no link to the pre-commit hook in .pre-commit-hooks.yaml.

    • Fix: add a "Use it in pre-commit" section with the hook id.

    Unverified, worth checking

    • Windows line endings may be rewritten on save. Settle it by formatting a CRLF file and inspecting it with od -c.

    Step 3: Positioning

    1. Tagline: Stable key order for JSON, so diffs show real changes
    2. GitHub About description: keep it. It says what the tool is in the first five words.
    3. Topics: add these. json, formatter, cli, python, pre-commit, code-review

    Key benefits: exits 1 on invalid JSON, so CI fails early (proof: tests/test_exit.py); ships a pre-commit hook (proof: .pre-commit-hooks.yaml).
    5. What I did not claim: no speed claim (no benchmark in the repository), no user numbers.

    Summary

    1. Correct the package name on README line 14.
    2. Add the six topics above.
    3. Add a pre-commit section to the README.

    I can apply the README fixes if you want. I have not edited anything. To add the topics, run:

    gh repo edit acme/tidyjson --add-topic json --add-topic formatter --add-topic cli --add-topic python --add-topic pre-commit --add-topic code-review
    

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Before you post your project anywhere, the README decides whether visitors stay. Orviqa Lite checks it the way a careful evaluator would.

    1. Health check: a script checks the README, licence, docs, examples, tests, CI and community files, plus the GitHub About description, topics and releases if the GitHub CLI is signed in. Anything it cannot see is reported as unknown, never as missing.
    2. README audit: eleven areas, including whether the quick start would run on a clean machine. Every finding quotes the line it is about and says what it costs you.
    3. Positioning: a tagline, About description and topics. Every benefit points to a file in your repo. If what you have is already good, it says keep it.

    It opens with a verdict: Ready, Ready after N fixes, or Not ready yet.

    It runs locally with your own agent, needs no API keys, and never edits a file or changes your GitHub repository without asking. The bundled script reads your repository's files and, through the GitHub CLI if installed, that repository's public metadata. Nothing is sent anywhere else.

    The last line of every answer links to Orviqa, the full product this skill comes from.

    Source: https://github.com/uoparaji/orviqa-lite

    Does not run your install or quick-start commands: the quick start is checked by reading the README and manifests, and anything that needs a run to confirm is listed as unverified. Without a signed-in GitHub CLI, the About description, topics, homepage and releases are reported as unknown.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 19 days ago

    • Free with an Agensi account

    Requires

    Python 3.8+
    GitHub CLI (gh), optional
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Listed19 days ago

    What's inside

    Frequently Asked Questions

    Orviqa Lite uses static analysis of your repository structure and a strict 'evidence-based' rubric to identify broken installation paths, missing project signals, and documentation gaps that cause developers to bounce.

    Python 3.8 or later for the bundled health-check script (standard library only), and an agent that can run shell commands. The GitHub CLI (gh) is optional: when it is signed in, the skill also reads the About description, topics, homepage and releases.

    Yes. The file checks, README audit and positioning work on any local repository. For GitHub metadata it uses your own signed-in GitHub CLI, so it can read a private repository you have access to. Nothing is sent to Orviqa.

    The SKILL.md instructions (health check, eleven-area README audit, positioning), the health-check script, a README, and an example audit. It is free.

    Unlike generic AI writers, this skill has a 'no-fabrication' rule that forces the agent to link every marketing claim or benefit directly to specific files, tests, or code blocks found in your repo.

    No. It reads the quick start and checks each command against what the recommended install leaves behind: the PATH, environment activation and missing steps. Anything it cannot confirm without running a command is listed under "Unverified, worth checking", with the command that would settle it.

    New in Technical Documentation