How does this skill avoid the 'hallucinations' common in other AI-generated READMEs?

Every command, feature and config row has to come from a file you supply (package.json, pyproject.toml, Cargo.toml, go.mod, Gemfile, Dockerfile, .env.example) or something you confirm. Anything it can't back up is left out or marked "Not stated". You still check the commands before you publish, and the README ends with a caveat saying so.

Which programming languages and package managers are supported?

Any project with a manifest your agent can read or you can paste: package.json (Node.js), pyproject.toml (Python), Cargo.toml (Rust), go.mod (Go) and Gemfile (Ruby), plus Dockerfile and .env.example. Other ecosystems work too if you paste the build and run commands.

Can I customize the README for different types of readers, such as developers vs. non-technical users?

Yes, you can specify if the documentation is for end-users, developers, or operators, and the skill will prioritize the relevant sections and technical depth accordingly.

Will the skill automatically generate 'fluff' content like generic badges or social media links?

No. It never adds badges, star or download counts, testimonials or "used by" logos unless you supply the facts behind them, and it only writes sections it has evidence for. It does not check that links work, so run a link checker before publishing.

What specific information do I need to provide for the skill to work effectively?

Four things: the project name, a one-line description, the audience (developers, end users, operators or mixed) and your manifest files or excerpts (for example package.json, Cargo.toml or .env.example). Your commands, status, license and current README are optional but make the result better.

What is included with my purchase of this AI agent skill?

The skill file (SKILL.md with the inputs table, 8-step method, honesty rules and README output template) plus a README, to install in Claude Code, Cursor, Codex or another SKILL.md agent. You buy it once. Cygnix Labs publishes improvements to this listing as new versions (current: v1.0.0, Oct 2026); how updates reach existing buyers follows Agensi's standard purchase terms.

Can it update an existing README instead of starting over?

Yes. Paste your current README with your manifests. It keeps what your files back up, rewrites thin or outdated sections, and flags commands or versions that no longer match.

Works with the AI Agents you already use

+21 more

+18 agents

README Generator: README.md from Your Real Project Files

Q: Why pay when free README generators exist?

Free generators scan your repo and fill a template. This one also writes an honest status line, keeps config to variable names only, and when you paste your current README it points out where the README and your files disagree. In the sample above it caught README examples that fail on the version npm installs, and a start command that needs a build step first.

README generator that writes README.md only from your real files: install/run commands from package.json or pyproject, a config table from .env.example, and "Not stated" instead of guesses.

Secure checkout via Stripe

Money-back guarantee

Converts to your local currency at checkout

Security scanned

README Generator: README.md from Your Real Project Files

Example session with this skill installed

project_name: Cygnix SEO MCP
one_liner: An MCP server with 5 SEO tools (keyword clusters, search intent, GSC cannibalization, title/meta checks, content briefs) that run locally with no API key.
audience: developers
manifests_or_excerpts: package.json, Dockerfile, smithery.yaml and src/index.ts from github.com/cygnixlabs-cmd/cygnix-seo-mcp (commit 4e7f271), pasted below
primary_commands: npm ci; npm test; npm start
status: 1.0.0, published on npm
license: MIT
repo_layout: src/index.ts, src/tools.ts, test/, docs/ci.yml.example, Dockerfile, server.json, smithery.yaml
output_depth: standard

Read your context and instructions
Compiled the readme generator

Cygnix SEO MCP

An MCP server with 5 SEO tools (keyword clusters, search intent, GSC cannibalization, title/meta checks, content briefs) that run locally with no API key.

Status: 1.0.0, published on npm · Audience: developers · License: MIT

Why this exists

SEO checks usually need a paid API or a SaaS login. These 5 tools work on data you paste (keyword lists, a Search Console CSV, your titles), so an MCP client can run them offline.

Features

Tool (from src/index.ts)	Input	What it does
`cluster_keywords`	keywords (lines or commas), optional `min_shared_terms` 1-3	Topical clusters with a pillar keyword each. Deterministic; no search volumes
`classify_search_intent`	keywords	Informational / commercial / transactional / navigational, from modifier rules
`find_keyword_cannibalization`	GSC CSV text (Query + Page), optional `min_impressions`	Queries where 2+ of your pages compete
`check_title_meta`	1-100 items {title, meta, keyword, url}	Length/pixel truncation, keyword placement, CTR hooks; 0-100 score with fixes
`content_brief_outline`	keyword, optional related keywords, audience, word_count 300-10000	Intent, title options, meta draft, H2 outline, questions, on-page checklist

Quick start

Prerequisites

Node.js 18+ (engines.node: ">=18")

Install and run from npm

npx -y cygnix-seo-mcp

(bin: cygnix-seo-mcp -> dist/index.js; stdio server, the same command smithery.yaml uses)

From source

npm ci
npm test     # runs tsc, then node --test test/
npm start    # node dist/index.js

Docker

docker build -t cygnix-seo-mcp .
docker run -i --rm cygnix-seo-mcp

(Dockerfile: node:20-alpine, npm ci && npm run build, entrypoint node dist/index.js)

Configuration

None observed in supplied materials: no .env.example, and smithery.yaml has an empty configSchema ("no configuration or API keys needed").

Project layout

src/index.ts: server and tool registration · src/tools.ts: tool logic
test/: tools.test.mjs (unit), e2e.test.mjs (starts the stdio server, lists and calls all 5 tools)
docs/ci.yml.example, server.json (MCP registry), smithery.yaml, Dockerfile

License

MIT (package.json license field and LICENSE file)

Caveat

This README was drafted from supplied project materials. Verify commands, versions, and claims with a human owner before publishing.

stdio server, the same command smithery.yaml

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

About this skill

Unlike free README generators that scan your repo and fill a template, this one checks every command against your manifests, writes an honest status line, keeps config to variable names only, and when you paste your current README it points out where the README and your files disagree. Anything it can't back up is marked "Not stated", never guessed.

Sample run (3 real public GitHub repos, files read 20 Oct 2026; every fact below comes from the repo's own files):

vercel/ms (commit 4ff48ce): package.json on main says v4.0.0, ESM-only, Node 20+, pnpm Flag: npm install ms still installs 2.1.3 (the npm "latest" tag), and the README's import { ms } from 'ms' fails on it (we ran it on Node 20: "Named export 'ms' not found") Wrote: Status "main = 4.0.0, not yet the npm latest (2.1.3)" · Prerequisites "Node.js 20+" · Tests "pnpm test" (the test script calls pnpm; the README says npm test)
pypa/sampleproject (commit 621e497): the current README has no install or run command at all Wrote: Quick start pip install sampleproject then sample (the [project.scripts] entry) · Python 3.9+ · Status "Alpha" (from the classifier) · License MIT Checked: pip installed 4.0.0 and sample printed "Call your main application code here" Tests: nox (default session: lint) and nox -s tests (from noxfile.py) · Configuration: none observed
excalidraw/excalidraw-room (commit 03ff435): the README has no description, config or production steps Config table from .env.development (names only): PORT (80, or 3002 in development, per src/index.ts) · CORS_ORIGIN (defaults to "*") Flag: yarn start runs dist/index.js, so yarn build must come first · the Dockerfile uses node:12-alpine and package.json has no engines field, so Node version = "Not stated" Flag: package.json name "excalidraw-portal" differs from the repo name, so it asks which one goes in the title Status: "Not stated" (no status given) · badges, stars, "used by" logos: none added

The full README from a 4th real run (our own MCP server, commit 4e7f271; npm ci + npm test passed 6/6) is in the Example below.

Works in Claude Code, Cursor and Codex · tested on real repos (the 3 above + the Example) · instructions-only skill, no scripts or API keys · never adds badges, star or download counts, or testimonials it can't back up · v1.0.0, updated Oct 2026 · questions answered via Agensi messages within 24h · Proof we ship real dev tools: https://github.com/cygnixlabs-cmd/cygnix-seo-mcp (open-source MCP server, on npm)

Give your agent the project name, a one-line purpose, the audience and your manifests (package.json, pyproject.toml, Cargo.toml, go.mod, Gemfile, Dockerfile, .env.example). You get a skimmable README.md that answers "what is it, who is it for, how do I run it, is it ready?" on the first screen.

What it does

Install and run commands taken from your manifests, never typed from memory
A config table from .env.example with variable names only, never values
An honest status line (idea, alpha, beta, stable, maintenance) and a known-limits note
Tone and section order for developers, end users or operators; lite or standard depth
Sections with no evidence are left out or marked "Not stated"

What's inside: SKILL.md (input table, 8-step method, honesty rules, the output template) + README.

Who it's for: maintainers shipping a first README, teams standardising READMEs across repos, anyone rewriting a thin or outdated README. Not for: API reference sites, legal text, or inventing features the code doesn't have.

Pairs well with (Cygnix Labs, on Agensi):

Write Commits Reviewers Approve: Conventional + Why (agensi.io/skills/write-commits-reviewers-approve-conventional-why): commit messages from the same evidence
PR Description Writer — Summary, Changes + Test Plan (agensi.io/skills/pr-description-writer-summary-changes-test-plan): the PR that ships the new README
Code Reviewer — Severity-Rated PR Review + Fix Plan (agensi.io/skills/code-reviewer-severity-rated-pr-review-fix-plan): review the code before you document it
Code Review Lite — Severity Checklist (free) (agensi.io/skills/code-review-lite-severity-checklist): a quick free check first
Find the Commit That Broke Your Code from Git History (agensi.io/skills/find-the-commit-that-broke-your-code-from-git-history): when the README's commands stop working

How to use

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

Money-back guarantee

Doesn't do what its page promises? You get your money back.

Trust & safety

Security scan

Security scan passed 1 month ago

Listed1 month ago

Updated7 days ago

More screenshots

README Generator: README.md from Your Real Project Files

README Generator: README.md from Your Real Project Files

Cygnix SEO MCP

Why this exists

Features

Quick start

Prerequisites

Install and run from npm

From source

Docker

Configuration

Project layout

License

Caveat

About this skill

Known limitations

How to use

Reviews

No reviews yet

Trust & safety

Permissions required

What's inside

Frequently Asked Questions

How does this skill avoid the 'hallucinations' common in other AI-generated READMEs?

Which programming languages and package managers are supported?

Can I customize the README for different types of readers, such as developers vs. non-technical users?

Will the skill automatically generate 'fluff' content like generic badges or social media links?

What specific information do I need to provide for the skill to work effectively?

What is included with my purchase of this AI agent skill?

Why pay when free README generators exist?

Can it update an existing README instead of starting over?