Loading connector details…
Loading connector details…
Choose a unique username to continue using AgentHotspot
by doobidoo • Uncategorized
Open-source persistent shared memory backend for multi-agent AI systems with fast retrieval and knowledge graph support.
Persistent memory shared across multiple runs and agents with fast 5ms retrieval.
Share causal knowledge graphs and maintain context without re-explaining.
A privacy-first, self-hosted memory backend compatible with various AI frameworks and clients.
MCP Memory Service provides a self-hosted, privacy-first memory backend that enables AI agents to store, share, and retrieve contextual knowledge and causal graphs in milliseconds without cloud lock-in or API costs. It supports a framework-agnostic REST API, local embeddings via ONNX, and real-time event streaming, making it ideal for multi-agent pipelines and integrations with popular AI tools like Claude and LangGraph. The service also supports enterprise features such as OAuth 2.0 and remote MCP for browser-based usage.
Fetch one section of the Project AI Memory (Strict Instructions). Call when the user's question or task maps to a single area (e.g. K8s → "Kubernetes Best Practices"). The returned text is project policy for that area: apply it; do not contradict it; on conflict, explain and recommend the project-approved approach. Section names: AI Security Rules, AI Role, Engineering Mindset, Cloud Platforms, Infrastructure as Code, Kubernetes Best Practices, Docker Best Practices, Database & DBA Best Practices, CI/CD Pipeline Standards, DevSecOps Security Rules, Observability & Monitoring, AI Engineering Best Practices, Code Quality Standards, Code Review Behavior, Performance Optimization, Response Style. Aliases: ai security, security rules, k8s, iac, terraform, dba, database, sql, security, devsecops, cicd, pipeline, monitoring, code quality, code review.
STRICT: You MUST call this before (1) implementing any security-sensitive change, (2) reviewing code that touches secrets, IAM, auth, networking, or compliance, (3) advising on security, vulnerabilities, or hardening. Returns the DevSecOps Security Rules from Project AI Memory (Strict Instructions). Authority: project policy—follow exactly; do not contradict. On conflict with user request, explain the conflict and recommend the project-approved approach.
STRICT: You MUST call this before writing, refactoring, or reviewing code so every suggestion and snippet conforms to project standards. Returns the Code Quality Standards from Project AI Memory (Strict Instructions). Authority: project policy— production-ready, SOLID, security, testing, style; treat as mandatory. Do not propose or approve code that violates them. On conflict with user request, explain and offer a compliant alternative.
Run `terraform init` then `terraform plan` in the given directory. Use for IaC review or before apply. path: Absolute or relative path to the Terraform project (must contain .tf files).
List Kubernetes pods. Uses kubernetes client; loads kubeconfig or in-cluster config (when running inside a pod). namespace: If set, list pods in that namespace; otherwise list all namespaces. Returns: pod names (and namespace if all), one per line.
Load the full Project AI Memory (Strict Instructions). Use when you need complete project context before a substantial task (new feature, architecture, or broad review). Authority: the content is project policy. You MUST apply it; do not suggest or implement anything that contradicts it. When the user requests something that conflicts, explain the conflict and recommend the project-approved approach. Sections cover: AI Role, Engineering Mindset, Cloud/IaC/K8s/Docker, Database & DBA, CI/CD, DevSecOps, Observability, AI Engineering, Code Quality, Code Review, Performance, Response Style.
Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.