AlexZio00

Sovereign Skills — Security skill for Claude Code

Security community

20 production-grade skills for AI coding agents — setup, scope, discipline, code review, security, session management, governance, ops, and quality audits (eval-leakage, doc-drift).

How to install Sovereign Skills

This entry records only its repository, not the path inside it, so there is no exact command to give. Open AlexZio00/claude-code-skills and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Sovereign Skills does

20 production-grade skills for AI coding agents — setup, scope, discipline, code review, security, session management, governance, ops, and quality audits (eval-leakage, doc-drift). Works with Claude Code, Codex, and Cursor.

Alternatives in Security

Skills Audit Report — Date: 2026-02-15 Auditor: Automated Skill Quality Audit Scope: Recently added skills in business-growth/, fina 5.3k ★
Cve MCP Server — Production-grade MCP server giving Claude 27 security intelligence tools across 21 APIs — CVE lookup, EPSS sco 1.2k ★
Perplexity Super Skills — Complete collection of 12 Perplexity Computer Super-Skills merging Perplexity + Claude Code capabilities acros 346 ★

README

🌐 **English** | [한국어](docs/README.ko.md) | [日本語](docs/README.ja.md) | [中文](docs/README.zh.md) | [Español](docs/README.es.md)

sovereign-skills v6.5.9

20 skills for the full Claude Code project lifecycle — from setup to daily workflow to code review to session management to governance. Each skill is useful standalone; the full sequence covers everything.

**What changed in v6.5.9:** Codex compatibility complete — all 20 skills now ship `agents/openai.yaml`. Previously 5 skills added in v6.3–v6.5 (`doc-drift`, `eval-leakage-audit`, `next-action`, `project-overview`, `skill-ops`) were missing Codex agent definitions. Installation section restructured (Option C: Codex/AGENTS.md, Option D: Cursor/other agents). No skill content changes — packaging-only release.

**What changed in v6.5.8:** Refinement release — no skills added or removed; a targeted delta port from the internal fork since v6.5.7 (6 of the 20 skills changed upstream in that window, plus a hand-diffed `code-autopsy` port). `doc-drift` (new Derivability signal — a hardcoded-but-code-reconstructable fact is flagged as structurally drift-prone even when currently correct), `integration-intake` (new Analogy-trap check — an "X does it this way, so should we" value claim needs three answered questions before it counts as evidence), `session-start` (new Query-conditional load rule for on-demand memory files — load only after confirming the conversation is actually about that topic, not on keyword overlap alone), `goal-lock` (new optional EVAL TYPE field for tasks that measure a skill/hook/gate's own reliability, a Silent Self-Correction anti-pattern, a First-Attempt Ledger recording the raw pre-fix DONE EVIDENCE run before any changes, an Early Self-Doubt Boundary note against premature task abandonment from misjudged remaining budget), `pre-push` (code-reviewer unavailability now falls back to an inline abbreviated review instead of an outright skip; new Step 3.5 Public-Mirror Scrub — a WARN-only push-ti