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.
/plugin marketplace add NovasPlace/CSMnpx agentmods add plugins/novasplace/csm/cross-session-memorygit clone --depth 1 https://github.com/NovasPlace/CSMGrade A, and why
cross-session-memory 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 yesterday.
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.
What it actually says
{
"name": "cross-session-memory",
"version": "1.0.0",
"description": "The complete self-hosted Cross-Session Memory runtime and lifecycle automation, native to Claude Code.",
"author": {
"name": "Donovan"
},
"homepage": "https://github.com/NovasPlace/CSM",
"repository": "https://github.com/NovasPlace/CSM",
"license": "MIT",
"keywords": [
"memory",
"claude-code",
"continuity",
"postgres",
"sqlite"
],
"commands": "./commands/",
"agents": "./agents/",
"skills": "./skills/",
"hooks": "./hooks/hooks.json",
"mcpServers": "./.mcp.json",
"interface": {
"displayName": "Cross-Session Memory",
"shortDescription": "Full CSM continuity runtime for Claude Code.",
"longDescription": "Memory, governance, living state, beliefs, self-model, AgentBook, checkpoints, context cache, goals, work ledger, onboarding, re-entry, compaction, and handoff automation — wired into Claude Code's native lifecycle hooks, slash commands, subagents, and skills.",
"developerName": "Donovan",
"category": "Productivity",
"capabilities": [
"Search",
"Write",
"Long-term memory",
"Lifecycle automation",
"Living state",
"AgentBook"
],
"defaultPrompt": [
"Search project memory before we start changing code.",
"Build a context brief for this task from prior sessions.",
"Show the current context pressure before we continue."
],
"brandColor": "#2563EB",
"websiteURL": "https://github.com/NovasPlace/CSM",
"privacyPolicyURL": "https://github.com/NovasPlace/CSM/blob/master/docs/DATA_PRIVACY_AND_LIFECYCLE.md"
}
}
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.
- yesterday First seen · 47 lines scan A fd2bfccdf445
cross-session-memory is a plugin published in the GitHub repository NovasPlace/CSM (41 stars, last pushed 8d ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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 2026-08-30.
Other plugins, from other repositories
graphjin-mcp
GraphJin MCP server plugin for Claude Code.
knowl
Persistent memory for Claude Code. Facts are typed, sourced, and retired when they change, so a query returns the current answer rather than every answer it has ever held. Local SQLite in your repository — no API key, nothing uploaded.
md-annotator
Browser-based Markdown annotator with text highlighting for Claude Code.
minimalist
Subtraction-first engineering for AI agents. The smallest correct change wins.
prompt-engineering-expert
Designing effective prompts: structure, few-shot, chain-of-thought, output formatting, and evaluation.
engineering-memlog
Engineering memory for AI coding agents — a structured cross-project log, a deliberately dumb CLI, and a Claude Code plugin that closes the read-side loop with auto-firing hooks.