---
title: "horizonlayer"
description: "Durable PostgreSQL knowledge and agent-friendly issue tracking for coding agents."
canonical: https://agentpluginsdirectory.com/plugins/horizonlayer
last-updated: 2026-10-21
---

# horizonlayer
Durable PostgreSQL knowledge and agent-friendly issue tracking for coding agents.
- Slug: horizonlayer
- Publisher: Kyle Mirich
- Repository: https://github.com/kyle-mirich/horizonlayer
- Manifest: plugins/horizonlayer/plugin.json
- Version: 0.1.2
- License: MIT
- Category (editorial): productivity
- Skills: 18 (code-review, codebase-design, diagnosing-bugs, domain-modeling, grill-with-docs, implement, improve-codebase-architecture, issues, knowledge, prototype, research, resolving-merge-conflicts, tdd, to-spec, to-tickets, triage, using-horizonlayer, wayfinder)
- MCP servers: 1 (horizonlayer)
- Stars: 1
- Repository created: 2026-03-20
- Repository last pushed: 2026-10-17
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/horizonlayer
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What horizonlayer does, in the publisher's words

HorizonLayer is a local PostgreSQL MCP server for coding-agent knowledge and issue tracking. PostgreSQL is canonical; Qdrant is an optional derived index for semantic retrieval.

> Status: Early release on the 0.x line. Bundled plugin manifests pin the exact package version for reproducible agent configurations; the commands below track @latest.

Coding agents lose context between sessions and coordinate poorly with each other. HorizonLayer gives them durable memory and shared work management through one MCP server:

- A compact MCP surface. One knowledge tool and one issues tool instead of dozens of narrow endpoints, so agents spend their context budget on work rather than tool descriptions. Module selection (HORIZONLAYER_MODULES) trims the catalog to what a project actually uses.
- Canonical data with a disposable retrieval index. PostgreSQL 17 stores pages, blocks, typed databases, rows, issues, comments, dependencies, and links. The optional Qdrant-backed RAG index is fully derived: drop it, and it rebuilds from canonical records.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/kyle-mirich/horizonlayer/HEAD/README.md

## Skills

- code-review: Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes, Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/PRD asked for?). Runs both reviews in parallel sub-agents and reports them…
- codebase-design: Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.
- diagnosing-bugs: Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.
- domain-modeling: Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.
- grill-with-docs: A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.
- implement: Implement a piece of work based on a spec or set of tickets.
- improve-codebase-architecture: Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.
- issues: Manage agent-friendly HorizonLayer Issues through MCP. Trigger when an agent needs Jira-style projects, searchable work items, exclusive assignment, comments, tags, subtasks, dependencies, readiness filtering, or explicit links between Issues and knowledge pages.
- knowledge: Use HorizonLayer as durable, workspace-scoped agent knowledge. Trigger when an agent needs to find prior context, save decisions or implementation knowledge, maintain pages and blocks, relate stored entities, or choose between canonical record search and citation-ready RAG retrieval.
- prototype: Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.
- research
- resolving-merge-conflicts
- tdd
- to-spec
- to-tickets
- triage
- using-horizonlayer
- wayfinder

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.

## MCP servers

- horizonlayer: transport: stdio; command: npx -y horizonlayer@0.1.1 mcp

Read from the plugin's own mcp.json. Environment variable names only, never values.