
Research
/Security News
77 Firefox Extensions Linked to Crypto Wallet and Credential Theft
Socket uncovered 77 linked Firefox extensions, including 40 that steal wallet secrets or credentials and 37 deceptive sports-score shells.
@akr1166org/rules-mcp
Advanced tools
AI Governance MCP — structured skills, knowledge, and evaluation for Cursor, Claude, VS Code, and Antigravity
AI Governance MCP: one source of truth for Markdown/MDC skills, knowledge, and evaluation — usable from Cursor, Claude, and any MCP-compatible agent.
Status: Phase 2 — AI governance (
0.3.0) — complete
Content format: MD / MDC only undercursor skills/(YAML rules deprecated)
Maintain Cursor-style skills as .md / .mdc under cursor skills/. Agents discover them through MCP (list_skills, get_skill, search_skills), enrich prompts, and pull context — without depending on YAML rule files.
See implementation.md for the full roadmap.
| Capability | Status |
|---|---|
Skill Engine (MD/MDC under cursor skills/) | ✅ |
| Knowledge / Context / Prompt engines | ✅ |
| Violation pipeline + plugin validators | ✅ (optional; no YAML seed rules) |
| Diff review / scorecard / explain | ✅ |
| Ranked knowledge search | ✅ |
| Full artifact suite + 7 agent profiles | ✅ |
| HTTP org server / IDE extension | Phase 3 |
ping, server_infolist_skills, get_skill, search_skills ← primary content APIlist_rules, get_rule, search_rules (legacy YAML path; empty by default)get_knowledge, search_knowledge, get_context, enrich_promptvalidate_code, validate_paths, review_diff, score_output, explain_violationgenerate_artifactscursor skills/
typescript-expert/SKILL.md
nestjs-best-practices/SKILL.md
nestjs-best-practices/rules/*.md
...
default, cursor, claude, continue, cline, gemini, chatgpt
npm install
npm run build
npm test
npm run rules:validate
Makes rules-mcp available on your PATH for Cursor, VS Code, Antigravity, Claude Desktop, etc.:
npm run link:global
rules-mcp version
Then point any MCP client at:
{
"command": "rules-mcp",
"args": ["serve"]
}
See docs/CONNECT.md for VS Code, Antigravity, and other clients.
Unlink later with npm run unlink:global.
Validate / score / review:
rules-mcp validate "const x: any = 1;" --file=src/x.ts
rules-mcp score "const x: any = 1;" --file=src/x.ts
rules-mcp review-diff --stdin < change.patch
Generate all artifacts:
rules-mcp generate --out=.
# or subset:
rules-mcp generate --target=cursor,agents,claude,styleguide,security --out=.
Start the MCP server:
rules-mcp serve
# or: npm start
Connect a client: docs/CONNECT.md
Natural-language prompts: docs/USER_PROMPTS.md
Evaluate Phase 2: docs/PHASE2_EVAL.md
rules/ # Enforceable governance rules (YAML)
cursor skills/ # Cursor agent skills (Markdown / MDC)
knowledge/ # Project knowledge
profiles/ # Per-agent profiles (7+)
src/ # MCP server + engines + validator plugins
examples/ # Client MCP configs
docs/ # Connect + phase eval
First phase/ # Phase 0 archive
| Variable | Default |
|---|---|
RULES_PATH | ./rules |
SKILLS_PATH | ./cursor skills |
KNOWLEDGE_PATH | ./knowledge |
PROFILES_PATH | ./profiles |
PROFILE | default |
LOG_LEVEL | info |
MIT
FAQs
AI Governance MCP — structured skills, knowledge, and evaluation for Cursor, Claude, VS Code, and Antigravity
We found that @akr1166org/rules-mcp demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?

Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.

Research
/Security News
Socket uncovered 77 linked Firefox extensions, including 40 that steal wallet secrets or credentials and 37 deceptive sports-score shells.

Security News
NIST disclosed an unreleased AI tool called V-etalon and opened a broad inquiry into NVD modernization after years of automation plans produced no public enrichment system.

Security News
In his AI Council 2026 talk, Feross Aboukhadijeh covers recent package compromises, vulnerability discovery, and a more automated security model.