dev-skills CLAUDE.md

dev-skills CLAUDE.md is an instructions file for coding agents from sungurerdim/dev-skills. It costs 4,980 tokens per session, scanned A, original, MIT.

Project instructions for AI coding assistants working with the dev-skills repository and its collection of software-development skills.

In plain words
What is it for?
Use them when maintaining that skills project, checking its structure and tests, following its Markdown-only conventions, or understanding its release and development process.
Why use it?
They tell the assistant how the repository is organised, what rules apply, how to validate edits, and which tools and platforms are supported.

Instructions file

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/sungurerdim/dev-skills/claude-md
Clone the repo
git clone --depth 1 https://github.com/sungurerdim/dev-skills

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for dev-skills CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/sungurerdim/dev-skills/claude-md.svg)](https://agentmods.dev/instructions/sungurerdim/dev-skills/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/sungurerdim/dev-skills/claude-md"><img src="https://agentmods.dev/badge/instructions/sungurerdim/dev-skills/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 4,980 This file is loaded in full into every session.
When invoked 4,980 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.04980 $0.04980
Opus 5 $0.02490 $0.02490
Sonnet 5 $0.00996 $0.00996
Haiku 4.5 $0.00498 $0.00498

Measured yesterday against content hash 071d70a78467, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

dev-skills CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 147 lines

How it starts

The opening of the file, as written. The whole thing — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.

dev-skills

Multi-phase AI coding assistant skills covering the full software lifecycle — gates, error recovery, and systematic mitigation of 17 known AI failure modes (W1–W11 universal + W12–W17 domain-specific) across Claude Code, OpenCode, Cursor, Copilot, Windsurf, and Aider.

Mission: enable a solo developer + AI to ship production-ready, high-quality products in any software project — encoding each domain's complete best practices (security, performance, privacy, accessibility, UX/UI, testing, compliance, operations) as executable gates, so quality does not depend on the developer's knowledge, attention, or available time. Target: the highest-quality, most practical skill set in its class — every critical point covered, nothing that doesn't earn its keep.

Meta

  • Status: active
  • Owner: sungurerdim
  • License: MIT
  • Stack: Markdown only — zero runtime dependencies
  • Skill count: 32
  • Tool support: Claude Code, OpenCode, Cursor, Copilot, Windsurf/Devin Desktop (June 2026 rebrand — newer builds prefer .devin/rules/), Aider (Agent Skills spec readers; OpenCode consumes ~/.claude/skills/ directly)
  • Releases: semver from v1.0.0 (2026-07-15); v2–v5 labels in docs/commits are spec-generation names, not release versions (see README § Versioning)
  • Active program: cross-host (spec-gen v5) — see docs/methodology/cross-host-program.md (2026-07-15 research: verified findings F1–F11, gaps G1–G6, program P0–P2, ds-rig)

Project Structure

Path Purpose
ds-<name>/ One directory per skill (ds-init, ds-fix, ds-review, ds-deploy, etc.)
SKILL-SPEC.md Authoritative skill format spec — every ds-* must conform
agents/ Shared agent definitions (ds-research-agent — worker for ds-research + ds-brief); install to the host agent dir, e.g. ~/.claude/agents/
docs/ methodology/ (research + program records, authoring checklist) and guide.html (interactive production-readiness guide); the former lifecycle guides were migrated into the owning skills' rules files in 2.0.0
core/ Shared runtime references every skill links to as ../core/<file>.md — principles, severity/score, findings + profile format, signal inventory, ask-exception list, checkpoint protocol, execution loop, report templates, secret patterns, toolchains, CRAAP, plus the software-best-practices.md catalog and the experience-rules.md XR pointer map. Shipped on every install; no SKILL.md inside
scripts/ The gate: quality.sh (entry point), check-consistency.sh (+ --self-test, --mutation-test), test-install.sh (installer round-trip)
install.sh · install.cmd The installer (bash + coreutils + git, no rsync) and its Windows launcher; --check, --status, --update, --uninstall
AGENTS.md Cross-host contributor instructions — the file Codex/Cursor/Copilot/Aider read; this file is the Claude-Code counterpart
.github/ Issue + PR templates only — no workflows, the gate is local by design

Read the full file on GitHub · 147 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday Changed · +10 lines · +983 tokens per session 071d70a78467
  2. 4d ago First seen · 137 lines · 3,997 tokens per session scan A c03ff90ebc55

Subscribe to this mod's changes

dev-skills CLAUDE.md is an instructions file published in the GitHub repository sungurerdim/dev-skills (1 stars, last pushed 2d ago), licensed MIT. It adds 4,980 tokens to every session, about $0.0249 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.