Minga — Security skill for Claude Code
Lean, security-first agent marketplace for Claude Code — a 3k-token core team (plan, build, verify, ship) plus 15 optional stack packs, with a built-in guard that blocks leaked secrets before they're.
How to install Minga
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open dmminga1983/minga and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Minga does
Lean, security-first agent marketplace for Claude Code — a 3k-token core team (plan, build, verify, ship) plus 15 optional stack packs, with a built-in guard that blocks leaked secrets before they're committed.
Alternatives in Security
- Security Fundamentals Pack — Core security skills and practices 1.6k ★
- Agent Skills Guard — 一款提供Agent Skills安全扫描和可视化管理的桌面应用 A desktop application that provides security scanning and visual management fo 389 ★
- Claude Leaked Files — Mirrored snapshot of Claude Code's source (exposed 2026-03-31) preserved for educational purposes, defensive s 256 ★
README
Minga
**The Claude Code team that plans, builds, verifies and ships your code — and never lets a secret slip out.**
Most agent marketplaces throw hundreds of skills at you and hope a few stick. Minga does the opposite: a tight 12-agent core (~3k tokens, always on) plus 15 optional packs you add only when you need them — one skill per stack, not one per framework. A guard hook blocks leaking API keys, tokens and `.env` files before they ever reach a commit, and a dedicated `verifier` agent double-checks finished work instead of grading its own homework. Free, MIT-licensed, zero network calls, zero telemetry.
[Русская версия →](README.ru.md)
/plugin marketplace add dmminga1983/minga
/plugin install minga@minga
/plugin install minga-python@minga # add the packs you need
Then type `/minga:start` — Minga asks what you want to do and points to one next step.
Why Minga
Big toolkits such as [ECC](https://github.com/affaan-m/ECC) ship hundreds of skills at once. Claude Code caps the skill listing at about 1% of the context window, so on overflow most descriptions are dropped and skills stop triggering — and the always-on cost is paid in every session. Minga makes different choices:
| ECC (as measured, Oct 2026) | Minga | |
|---|---|---|
| Always-on context cost | ~44.8k tokens | ~3.0k tokens for the core; ~180 per stack pack |
| Structure | one big plugin, ~68 agents, ~290 skills | core + 15 optional packs; one skill per stack with reference files read on demand |
| Security | configuration scanner | a guard that is on from install: blocks commits and pushes containing keys, tokens or .env/key files |
| Independent checking | — | a verifier agent that never grades its own work |
| Languages | English | agents reply in the user's language; skills trigger in Russian and English |
| Network and money | paid tiers, referral links | no network calls, no paid actions, no ads, no telemetry |
| Dependencies | — | none; Node ≥ 18 only fo |
Related Skills
Artel
Lean, security-first agent marketplace for Claude Code — a 3k-token core team (plan, build, verify, ship) plus
Claude Clean Context
Keep Claude Code's always-loaded context lean: deterministic memory guard-rail distiller + skills audit. Never
Agent Overwatch
A deterministic guard for AI coding agents — checks each tool call before it runs (allow/ask/deny) and blocks
Avant Garde
Think before you build. Guard before you ship. A Claude skill: 6 foundation docs + a security gate before any
Rule Autopsy
Audits a CLAUDE.md or rules folder and reports how many blocks claim an effect without a reproducible receipt,
Wp Plugin Skill
Build WordPress plugins with AI that don't ship vulnerabilities. A Claude skill covering all 18 Plugin Handboo
Related Agents
Security Officer
OWASP + STRIDE pass over a change or surface. Fails closed — unresolved risk blocks ship. Spawn for anything t
Secret Purist
The paranoid sentinel of credential security. Use this agent to scan codebases and git history for leaked secr
Opensource Sanitizer
Pre-public-push safety net. Scans a diff or working tree for leaked secrets, PII, internal references, interna