saqlainxoas/agentic-coding-demo — explained in plain English
Analysis updated 2026-05-18
Copy the create-spec, test-feature, and ship skills into your own project for a controlled AI coding workflow
Review a generated test-feature report before approving code changes from an AI agent
Learn how to structure CLAUDE.md and AGENTS.md files to map a codebase for AI tools
| saqlainxoas/agentic-coding-demo | 0xallam/my-recipe | 0xhassaan/nn-from-scratch | |
|---|---|---|---|
| Stars | 0 | — | 0 |
| Language | Python | Python | Python |
| Last pushed | — | 2022-11-22 | — |
| Maintenance | — | Dormant | — |
| Setup difficulty | easy | moderate | moderate |
| Complexity | 2/5 | 2/5 | 4/5 |
| Audience | developer | general | developer |
Figures from each repo's GitHub metadata at analysis time.
Just copies markdown skill files into your project, no dependencies to install.
agentic-coding-demo is a small example repository showing a structured workflow for using AI coding assistants like Claude Code, Codex, or Cursor in a controlled way, rather than letting them make changes freely. The approach relies on three kinds of files: a CLAUDE.md or AGENTS.md file that maps out the codebase for the AI, SKILL.md files that store repeatable workflows as plain markdown instructions, and a docs/specs folder that keeps one written specification per feature. The overall loop has four steps. First, the create-spec skill takes a plain English feature description, checks that your working directory is clean, creates a new branch off an up to date main branch, and writes a short specification file covering the problem, the proposed solution, the affected areas, and the criteria for calling it done. Second, you enter Plan Mode, where the AI reads the specification along with the repository map and proposes an implementation plan for you to approve before it writes any code. Third, the test-feature skill reads the specification and the code changes made against the main branch, writes and runs focused tests for the new behavior, reviews the changed lines for bugs or missing validation, and returns a short report with a ship or do not ship verdict, without ever editing source files itself. Fourth, the ship skill commits any remaining changes, pushes the branch, and opens a pull request with a title, summary, and verification steps generated from the specification and the diff, refusing to run on anything other than a feature, fix, or chore branch so it cannot accidentally merge into a long lived branch. The repository is meant to be copied into other projects, since the skill files are plain markdown that any compatible AI coding tool can read. The example project included is a basic FastAPI application used only as a realistic canvas to demonstrate the workflow, with three real pull requests in the repository's history showing what each skill actually produces. The README does not state a license.
A demo repository showing a spec driven workflow for AI coding agents using SKILL.md, CLAUDE.md, and AGENTS.md files.
Mainly Python. The stack also includes Python, FastAPI, Markdown.
The README does not state a license.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
double-check against the repo, no cap.