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.
git clone --depth 1 https://github.com/beingmartinbmc/jambavanWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/plugins/beingmartinbmc/jambavan/jambavan)<a href="https://agentmods.dev/plugins/beingmartinbmc/jambavan/jambavan"><img src="https://agentmods.dev/badge/plugins/beingmartinbmc/jambavan/jambavan/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/plugins/beingmartinbmc/jambavan/jambavan"><img src="https://agentmods.dev/badge/plugins/beingmartinbmc/jambavan/jambavan.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
jambavan 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 1mo 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.
What it actually says
{
"name": "jambavan",
"description": "Local-first codebase memory for coding agents: AST-aware context, durable decisions, failure records, review packs, and a local graph GUI. Twenty tools are advertised by default; write_file, patch_file, prompt compression, and bash are opt-in. Discipline protocols ship as MCP prompts rather than tools. Pre-2.0 tool names still resolve. No LLM calls, telemetry, or code upload; repository indexes and the global memory archive stay local.",
"version": "2.0.2",
"author": {
"name": "beingmartinbmc"
},
"homepage": "https://github.com/beingmartinbmc/jambavan#readme",
"repository": "https://github.com/beingmartinbmc/jambavan",
"license": "MIT",
"keywords": [
"mcp",
"model-context-protocol",
"code-index",
"tree-sitter",
"developer-tools",
"mcp-server",
"llm-memory",
"coding-agents",
"persistent-memory",
"prompt-compression",
"local-first",
"review-automation",
"agent-memory",
"claude-code",
"cursor",
"codex"
],
"mcpServers": {
"jambavan": {
"command": "npx",
"args": [
"-y",
"jambavan"
]
}
}
}
What it installs
The manifest is a name and a version. 5 skills travel with it, and installing the plugin installs all of them — 292 tokens a session between them. Each is measured on its own page, and each can be installed alone.
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.
- 1mo ago First seen · 39 lines scan A 26d665fa2fd7
jambavan is a plugin published in the GitHub repository beingmartinbmc/jambavan (2 stars, last pushed 3d 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 2026-08-31.
Other plugins, from other repositories
deja
deja-vu memory for Gemini CLI: the coding sessions already on your disk — Gemini CLI, Claude Code, Codex, Cursor and thirty-eight more — searchable and recalled, including work from before it was installed.
ownmem
Git-native project memory for AI coding agents, with deterministic local recall, evidence gates, and layered delivery that quotes, points, or abstains.
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.
causal-memory
Persistent causal memory for Claude Code — decisions, outcomes, and lessons that survive compaction. Same-context counterfactuals (natural experiments), a falsifiable prediction ledger, and an activation skill that teaches when to recall and record.
memex
Zettelkasten-based agent memory system with bidirectional links.
yantrikdb
Cognitive memory for AI agents — semantic recall, a knowledge graph, contradiction detection and procedural learning. Local-first: one SQLite file, no account, no telemetry.