agentmods.dev

gnosys AGENTS.md

A set of instructions for working on Gnosys, an open-source system that stores an AI agent’s long-term memories in a SQLite database and provides them through tools and a command-line program.

In plain words
What is it for?
Use it when changing Gnosys, checking its build and tests, or working with its memory database, tool server, command-line interface, and search system.
Why use it?
It gives coding agents the project’s architecture, rules, and commands so they can work consistently without rebuilding that context each session.

Instructions file for CodexOpenCode

4 16d ago A scan 1,159 tokens original MIT
View the source file proticom/gnosys
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/proticom/gnosys/agents-md
Clone the repo
git clone --depth 1 https://github.com/proticom/gnosys

Made for: Codex, OpenCode.

Per session 1,159 This file is loaded in full into every session.
When invoked 1,159 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.01159 $0.01159
Opus 5 $0.00580 $0.00580
Sonnet 5 $0.00232 $0.00232
Haiku 4.5 $0.00116 $0.00116

Measured 2d ago against content hash 82144ade0aa3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

gnosys AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Instructions for AI agents working in the Gnosys codebase.

What is Gnosys

Gnosys is an open-source persistent memory system for AI agents. It stores structured memories in a central SQLite database and exposes them via MCP (Model Context Protocol) tools and a CLI. Agents use Gnosys to remember decisions, architecture, project status, and context across sessions.

Quick Reference

npm run build          # tsc -> dist/ (tsconfig.build.json — excludes tests)
npm test               # vitest run (1700+ tests, all should pass)
npm run typecheck      # tsc --noEmit (covers tests too — CI-gated)
npm run lint           # biome (zero warnings — CI-gated)
npm run knip           # dead code / unused exports (zero findings — CI-gated)

Architecture

  • Entry points: src/index.ts (MCP server, 50+ tools), src/cli.ts (CLI, 30+ commands)
  • DB-only: Central SQLite at ~/.gnosys/gnosys.db is the sole source of truth. No markdown files.
  • Search: FTS5 keyword -> semantic embeddings -> hybrid RRF -> federated cross-project
  • Web KB: gnosys/web subpath export for serverless chatbots (zero native deps at runtime)
  • Portfolio: src/lib/portfolio.ts + portfolioHtml.ts for cross-project status dashboard
  • Multi-machine sync (v5.3.0): src/lib/remote.ts syncs local cache with remote (NAS) DB

Key Rules

  1. DB-first lookups: When resolving a memory by ID (e.g. road-007), always check centralDb.getMemory(id) before falling back to the legacy file resolver.
  2. No markdown writes: All memory writes go to SQLite only. Markdown is generated on-demand via gnosys export.
  3. TypeScript strict: strict: true in tsconfig. Fix all type errors before committing.
  4. Test before commit: Run npm test — the full suite must pass (1700+ tests in src/test/). CI also gates tsc --noEmit, biome lint, and knip.
  5. Path quoting: The project may live in a path with spaces (iCloud). Always quote paths in shell commands.
  6. Web subpath isolation: src/lib/staticSearch.ts must not import (only import type) from modules that depend on better-sqlite3 or any native addon.
  7. CLI-first, MCP-second: New features are CLI commands first; MCP tools are thin wrappers around CLI logic.

Read the full file on GitHub · 89 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 89 lines · 1,159 tokens per session scan A 82144ade0aa3

Subscribe to this mod's changes

gnosys AGENTS.md is an instructions file published in the GitHub repository proticom/gnosys (4 stars, last pushed 16d ago), licensed MIT. It adds 1,159 tokens to every session, about $0.0058 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

LLMMemory and contextaiai-agentsai-toolsdeveloper-toolsknowledge-management
Related

Other instructions, from other repositories

raytsystem-public-os CLAUDE.md

Instructions for romarayt/raytsystem-public-os, covering compatibility pointer and documentation synchronization.

romarayt/raytsystem-public-os · 611 tokens

raytsystem-public-os AGENTS.md

Instructions for romarayt/raytsystem-public-os, covering raytsystem — agent routing, invariants, commands, skill routing and documentation.

romarayt/raytsystem-public-os · 684 tokens

harness-kit CLAUDE.md

Instructions for harnessprotocol/harness-kit, covering harness-kit, repo structure, adding a new plugin, 1. create the plugin directory and 2. write the plugin manifest.

harnessprotocol/harness-kit · 1,579 tokens

obsidian-pkm-plugin CLAUDE.md

Instructions for AdrianV101/obsidian-pkm-plugin, covering pkm: 01-projects/obsidian-mcp, claude.md, project overview, commands and install dependencies.

AdrianV101/obsidian-pkm-plugin · 4,656 tokens

brain-starter CLAUDE.md

Instructions for michaeljauk/brain-starter, covering brain repo - claude instructions, purpose, not set up yet?, file placement rules and directory guide.

michaeljauk/brain-starter · 715 tokens

Metis CLAUDE.md

Claude Code instructions for SVerITG/Metis, covering metis research cortex — claude code configuration, core operating rules, contextual discovery — actively guide the user to features, how to invoke agents and how invocation works.

SVerITG/Metis · 3,589 tokens