mxm-pack-l3-2-legal

The behavioral intelligence layer for Claude Code. 91 specialist agents across 7 executive offices, 52 skill domains, 50 slash commands, 86 frameworks (behavioral core + governance), 14 compliance frameworks, 9 MCPs (95 tools) including NotebookLM research synthesis, license-gated MCP middleware. 90-day Trial of all 14 packs default at install. Core tier is $0 forever; paid packs install from this same marketplace after purchase. BSL 1.1 (Apache 2.0 on 2030-04-21).

200 skills · 9 MCP servers

Audit trail, confidence tagging, and ethical checks on every AI decision. Primary framework: Prospect Theory (Kahneman & Tversky, 1979). MOAT-01.

1 skill

Cross-session memory that resumes your project where you left it. Primary framework: Cognitive Load Theory (Sweller, 1988). MOAT-02.

1 skill

Ten drift classes scanned continuously. Catches regressions before they ship. Primary framework: Signal Detection Theory (Green & Swets, 1966). MOAT-03.

1 skill

14 regulated-industry frameworks enforced at the MCP layer. Audit-ready on demand. Primary framework: COM-B (Michie et al., 2011). MOAT-04.

1 skill

Brand voice locked across AI outputs. 15 cinematic styles. Design that doesn't read as AI. Primary framework: Dual Process Theory (Kahneman, 2011). MOAT-05.

1 skill

Install the "Architecture & Domain Modeling" workflow plugin from Agentic Awesome Skills.

13 skills

Portfolio monitoring, valuations and analytics for private-capital GP teams using the Chronograph connector: portfolio-company one-pagers, fund quarterly review packs, markup/markdown briefs, TVPI attribution and budget-vs-actuals. Requires a Chronograph account and the Chronograph connector.

5 skills · 6 agents

HR policy documentation, legal compliance templates (GDPR/SOC2/HIPAA), employment contracts, and regulatory documentation

2 skills · 2 agents

Briefing pack before every client meeting

4 skills · 1 agent

Ready-to-run workflows for financial advisors in Claude Cowork, drawing live data from the tools their firm already uses through third-party connectors. The plugin holds no client data, and the advisor approves any write back to an external system. Claude gathers, organizes, summarizes, compares and presents information, while the advisor exercises judgment, makes the recommendation and gives the advice. For financial professional use only; not investment, legal, or tax advice.

8 skills · 6 agents · 36 MCP servers

A four-tier memory ladder for Claude Code where promotion is earned by recurrence, not asserted by confidence. L0 raw transcripts are never injected; L1 candidate atoms are gitignored and recalled on lexical relevance per prompt; L2 project context loads each session start; L3 stable persona is always loaded. L1 to L2 needs three distinct sessions spanning two distinct calendar days (a stated claim needs two; a verified claim is the only single-observation path); L2 to L3 needs two projects and thirty days. Two gates refuse rather than guess: a claim altered by the redaction pass never promotes on evidence alone, and a contradicted claim freezes until a human resolves it. Three hooks run the loop unattended and every one fails open. Nothing reaches a committed CLAUDE.md without an explicit human adopt. Ships 5 stdlib scripts, 3 hooks, a cs-memory-curator agent, /cs-memory, a JSON schema, and a 69-check validator. Concept from TencentCloud/TencentDB-Agent-Memory (MIT); no upstream code included.

1 skill · 1 command · 1 agent · hooks