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.
npx agentmods add instructions/mk-986123/server-memory/agents-mdgit clone --depth 1 https://github.com/MK-986123/server-memoryWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00808 | $0.00808 |
| Opus 5 | $0.00404 | $0.00404 |
| Sonnet 5 | $0.00162 | $0.00162 |
| Haiku 4.5 | $0.00081 | $0.00081 |
Grade C, and why
server-memory AGENTS.md scanned grade C with 1 finding 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
rm -rf dist build How it starts
The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository instructions for coding agents
These instructions apply to AI coding agents and automated development tools working in this repository.
Objective
Preserve server-memory as a local-first MCP memory server with explicit storage, bounded recall, workspace and global scope separation, and predictable stdio behavior.
Do not introduce hosted services, external credentials, telemetry, or mandatory network dependencies into the core package without explicit maintainer approval.
Initial setup
- Confirm Python 3.10 or newer.
- Confirm SQLite FTS5 support.
- Create an isolated virtual environment.
- Install the editable project with development dependencies.
Linux and macOS:
python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -e ".[dev]"
python -c "import sqlite3; c=sqlite3.connect(':memory:'); c.execute('CREATE VIRTUAL TABLE t USING fts5(content)'); c.close(); print('FTS5 available')"
Windows PowerShell:
py -m venv .venv
.venv\Scripts\Activate.ps1
python -m pip install --upgrade pip
python -m pip install -e ".[dev]"
python -c "import sqlite3; c=sqlite3.connect(':memory:'); c.execute('CREATE VIRTUAL TABLE t USING fts5(content)'); c.close(); print('FTS5 available')"
Install optional embeddings only when the task requires them:
python -m pip install -e ".[dev,embeddings]"
Safe test environment
Never use a real user memory database for tests or exploratory commands.
Set an isolated database path and disable global memory unless the task specifically tests global scope behavior:
export MEMORY_DB_PATH="$PWD/.tmp/agent-memory.db"
export MEMORY_GLOBAL_DB_ENABLED=false
PowerShell:
$env:MEMORY_DB_PATH = "$PWD\.tmp\agent-memory.db"
$env:MEMORY_GLOBAL_DB_ENABLED = "false"
Do not commit .tmp, SQLite databases, exports, backups, token files, model caches, virtual environments, or build artifacts.
Before editing
- Read
README.md,CONTRIBUTING.md, and the files directly involved in the task. - Inspect existing tests before changing behavior.
- Preserve public tool names, parameter shapes, scope semantics, and JSON-RPC behavior unless the task explicitly requires a breaking change.
- Keep stdio stdout protocol-clean. Send diagnostics to stderr or logging.
- Keep destructive operations explicit and scoped. Do not weaken rejection of
scope="all"for destructive tools. - Do not add performance claims without reproducible raw results.
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.
- 2d ago First seen · 116 lines · 808 tokens per session scan C d4ed896fc604
server-memory AGENTS.md is an instructions file published in the GitHub repository MK-986123/server-memory (1 stars, last pushed 24d ago), licensed MIT. It adds 808 tokens to every session, about $0.0040 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
m_flow AGENTS.md
AGENTS.md instructions for FlowElement-xinliuyuansu/m_flow, covering m-flow — developer & agent reference, 1. repository map, extension points, 2. local development and python backend (requires python 3.10 – 3.13).
Meterless AGENTS.md
Instructions for Meterless/Meterless, covering agent instructions for meterless, routing table, isolation rule, repo-wide rules and non-coding agents.
open-bridge AGENTS.md
Instructions for bks-lab/open-bridge, covering the bridge — agent instructions, required reading, session start detection (automatic), theme and agents.
second-brain AGENTS.md
Instructions for LindaHaviv/second-brain, covering instructions for ai coding agents, what to read first, the rules (non-negotiable), common tasks, the sanctioned way and enforcement, not just instructions.
cairn CLAUDE.md
Instructions for zilet/cairn, covering claude.md, what cairn is, commands, hard requirements and the traps that actually bite.
Agent-Context CLAUDE.md
Instructions for lx-wnk/Agent-Context, covering claude.md, project overview, commands, format all files and check formatting (ci-style).