client-memory

Maintain persistent client records with timestamped status changes and GDPR retention.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/stanz-stanz/heimdall --skill client-memory
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: client-memory
Source: https://github.com/stanz-stanz/heimdall/tree/main/.claude/agents/client-memory
Command: npx skills add https://github.com/stanz-stanz/heimdall --skill client-memory

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Centralizes and maintains authoritative persistent state for every onboarded client so teams can reliably track technology baselines, scan results, finding lifecycle, remediation progress, message delivery, and offboarding actions without data fragmentation or conflicting updates.

Core Features & Use Cases

  • Authoritative Client Profiles: Create and store client profile.json records with contact details, technical context, technology stack, tier, and scan schedule.
  • Historical Scan & Finding Tracking: Maintain history.json with scan records, finding status history, follow-ups, and message delivery metadata to support audits and decision-making.
  • Structured Update Processing: Accept and process structured update requests from other agents (e.g., post-scan findings, message receipts) while enforcing single-writer semantics and GDPR-aware retention.
  • Escalation & Offboarding: Flag stale/unresolved findings for escalation and perform offboarding workflows that mark status, archive anonymized analytics, or delete data per retention policy.
  • Use Case Example: When a new scan detects a vulnerability, update or create the finding entry, increment follow-ups, and expose the status to Message Composer and Project Coordinator for follow-up actions.

Quick Start

Create a profile and history for a new client "Restaurant Nordlys" with WordPress, weekly scans, and onboarding metadata so the agent can accept structured updates after scans.

Frequently Asked Questions about client-memory

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I maintain a single source of truth for client state across multiple agents?

To maintain a single source of truth for client state, use a centralized memory system that enforces single-writer semantics for client data directories. This ensures all agents submit structured updates to one authoritative profile and history record, preventing fragmented or conflicting updates.

What is the best way to track vulnerability remediation progress and scan history for clients?

The best way to track vulnerability remediation progress is to maintain a persistent history file with timestamped finding statuses, scan records, and follow-up metadata. This allows you to increment follow-ups, update finding entries, and expose statuses to other agents for action.

How does GDPR retention and auditability work with persistent client profiles?

GDPR retention and auditability work by enforcing strict data retention policies during offboarding workflows. The system archives anonymized analytics or deletes client profiles and history records per policy, while maintaining timestamped status changes to support audits.

Can I process structured update requests from other agents without write conflicts?

Yes, you can process structured update requests without write conflicts by requiring exclusive write authority to the client data directory. This single-writer approach processes incoming updates from other agents sequentially to prevent data corruption.

How do I handle client offboarding and escalate unresolved security findings?

To handle client offboarding and escalate unresolved findings, flag stale findings for escalation and execute an offboarding workflow. This marks the client status, archives anonymized analytics, or deletes data per your configured retention policy.