Harden Codebase Properly — Security skill for Claude Code
A Claude Code skill: measure an existing web app, repair it, and keep it at a verifiable production bar — gates, data and security, performance, interface, user communication, tests, documentation, ha.
How to install Harden Codebase Properly
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open finzidev/harden-codebase-properly and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Harden Codebase Properly does
A Claude Code skill: measure an existing web app, repair it, and keep it at a verifiable production bar — gates, data and security, performance, interface, user communication, tests, documentation, handover.
Alternatives in Security
- Harden — Run the harden skill to perform a systematic project hardening audit 1k ★
- Pareto Mac — by ParetoSecurity - Serves as development guide for Mac security audit tool with build instructions, contribut 431 ★
- Openclaw Ops — OpenClaw operations skill with health checks, repair scripts, watchdogs, update triage, and security scans 225 ★
README
harden-codebase-properly
A skill for Claude Code (and agents compatible with the `SKILL.md` format) that takes an existing web app or back-office, measures it, repairs it, and brings it up to a defined, verifiable production bar on everything that is not code structure: quality gates and tests that tell the truth, database permissions and secrets, migrations and environments, fast database loads, interface behaviour, readable errors for the user, documentation that is measured, client handover. A systematic, eight-phase procedure run through subagents: the orchestrator reasons, dispatches, reads, decides, and verifies; it does not write code.
The structural half (layers with an enforced dependency rule, bounded contexts and domain events, one implementation per concept, deep modules, files split by invariant, dead code, the table shell and the field registry) lives in the sibling skill `design-software-properly`. The two skills share the same method (briefs, verifier, decision records, evidence ledger) and the same `docs/decisions/` register; this one writes `docs/remediation/`, the sibling writes `docs/design/`.
Install
git clone https://github.com/finzidev/harden-codebase-properly.git ~/.claude/skills/harden-codebase-properly
Then, in a session open on the project you want to bring to production quality:
/harden-codebase-properly diagnose
Modes: `diagnose` (Phase 0 and the ordered plan; its plan question is the only mandatory user checkpoint of a remediation run), `plan`, `run `, `review` (a diff against the target state), `maintain` (guard set, change-time protocol, drift check), `status` (resume). With no argument the skill reads `docs/remediation/status.md` and continues from the first step without evidence.
Structure
| File | Role |
|---|---|
SKILL.md |
the router: modes, the six measurable conditions, non-negotiables, excuses and reality, red flags, file map |
modes/ |
one file per mode: diagnose (Phase 0, the |
Related Skills
CLI Code Skills
Production-ready Claude Code skills — audit code quality, forge design docs, generate documentation, automate
Interface Design Audit
Check existing code against your design system for spacing, depth, color, and pattern violations.
Documenter
Audit and improve open-source documentation against best-practice rubrics (Diátaxis, Good Docs Project, Standa
Cleanup Docs
Audit and clean up a repo's Claude documentation — measure the always-on tier first (CLAUDE.md + nested + .cla
App Audit
Grade an existing app against the Mobify Studio House KB and Axiom standards — fan out the specialist auditors
Walkthrough Nt
Role-driven browser walkthrough — identify each user role, drive the running app through their journeys in a r
Related Agents
Autoresearch Orchestrator
Use this agent to run an autoresearch experiment session end-to-end — setup, baseline, and a batch of edit→mea
Fec Debugger
Front-end diagnostic and repair subagent: Handle build failures, runtime errors, UI exceptions, and interface
UX Bar Raiser
Holds the UX and UI bar for an app, web or terminal. Audits the running app against Nielsen's heuristics, WCAG