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 Metabuilder-Labs/tokenjamnpx agentmods add plugins/metabuilder-labs/tokenjam/marketplacegit clone --depth 1 https://github.com/Metabuilder-Labs/tokenjamGrade A, and why
tokenjam 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.
What it actually says
{
"name": "tokenjam",
"owner": {
"name": "Metabuilder Labs",
"url": "https://tokenjam.dev"
},
"description": "Claude Code plugin marketplace for tokenjam: local-first token-cost observability and fixes.",
"plugins": [
{
"name": "tokenjam",
"source": "./plugin",
"description": "Local-first observability and cost control for Claude Code agents: a zero-token statusline showing context re-read %, savings/optimize reports, and a health check."
}
]
}
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 · 16 lines scan A 5bd7dea5c8ee
tokenjam is a plugin published in the GitHub repository Metabuilder-Labs/tokenjam (106 stars, last pushed 4d 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
agentsys
26 specialized plugins for AI workflow automation - task orchestration, PR workflow, slop detection, code review, drift detection, enhancement analysis, documentation sync, unified static analysis, durable memory, negative behavior memory, skill and system prompt curation, perf investigations, topic research, agent…
Arbor
Run Arbor's hypothesis-tree research workflow inside Claude Code using your own model — no API key. Bundles the Arbor Agent Skill suite and registers the keyless arbor mcp tools (Idea Tree, evaluation, worktrees, guarded merges, reports).
agent-harness-generator
MetaHarness — mint a custom AI agent harness from any repo. Paste a GitHub URL or start blank; output runs on Claude Code, OpenAI Codex, pi.dev, Hermes, OpenClaw, or RVM. WASM kernel (Rust → wasm-bindgen + NAPI-RS), 6 hosts, witness-signed Ed25519 releases, GCP Secret Manager validation. Browser Studio + npx metaharne.
metaharness
MetaHarness Marketplace: mint a custom, branded AI agent harness from any repo — runs on Claude Code, OpenAI Codex, pi.dev, Hermes, OpenClaw, RVM, Copilot, OpenCode, or GitHub Actions. WASM kernel, witness-signed releases, npx CLI.
kimi-k3-harness
Repo-aware coding agent harness for kimi-k3-in-c — Kimi K3 (2.78T-param) inference in portable C99: one CPU, 8 GB RAM, no BLAS, no framework, no GPU.
amp-plugin
AI Maestro AMP channel — injects inbox messages into a running Claude Code session so an idle agent wakes and processes them, with no tmux keystrokes.