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 CodeShuX/tokenwisenpx agentmods add plugins/codeshux/tokenwise/marketplacegit clone --depth 1 https://github.com/CodeShuX/tokenwiseGrade A, and why
tokenwise 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": "tokenwise",
"owner": {
"name": "CodeShuX",
"url": "https://github.com/CodeShuX"
},
"metadata": {
"description": "TokenWise — measurement-driven model router for Claude Code. Classifies every task by type and routes it automatically: mechanical→Haiku, execution→Sonnet, review→Opus, planning→Fable. Measures real dollars saved per session.",
"version": "0.2.0"
},
"plugins": [
{
"name": "tokenwise",
"source": "./",
"description": "Measurement-driven model router for Claude Code. Classifies every subtask by type and routes it automatically — mechanical work to Haiku, execution to Sonnet, review to Opus, large planning to Fable — with no per-task prompts, logs token usage per task class to a local NDJSON, and proves savings via on-demand A/B tests. Auto-detects whether routing/env-var primitives work on your Claude Code build before configuring. Exposes /tokenwise:install, /tokenwise:report, /tokenwise:summary, /tokenwise:ab, /tokenwise:undo.",
"version": "0.2.0",
"author": {
"name": "CodeShuX",
"url": "https://github.com/CodeShuX"
},
"homepage": "https://github.com/CodeShuX/tokenwise",
"repository": "https://github.com/CodeShuX/tokenwise",
"license": "MIT",
"category": "productivity",
"tags": [
"claude-code",
"anthropic",
"model-routing",
"token-optimization",
"cost-reduction",
"haiku",
"sonnet",
"opus",
"fable",
"subagents",
"measurement"
]
}
]
}
What ships with it
1 file beside marketplace.json in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 41 lines scan A b0e9723596ad
tokenwise is a plugin published in the GitHub repository CodeShuX/tokenwise (3 stars, last pushed 6d 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
frontend-slides
Zero-dependency HTML presentation generator with safe presets, bold template-library styles, PPT conversion, and anti-AI-slop design philosophy.
zero-slop
Zero Slop is a skill, not an AI model. Claude or another compatible model performs the editorial roles; local tools score the writing and protect facts, then the assistant and tools verify the final text.
li-evan
Plugin marketplace listing 1 plugin: bloom.
canadian-finance-planner
A comprehensive personal finance planning skill for Canadians. Conducts financial interviews, builds budgets and action plans, generates interactive dashboards, and provides ongoing coaching — all tailored to Canadian tax rules, registered accounts (RRSP, TFSA, RESP, FHSA), government benefits (CCB, CESG, CPP, OAS)…
linkedin-article-publisher
Publish Markdown articles to LinkedIn Articles.
claude-dnd-skill
Dungeon Master assistant for running persistent D&D 5e campaigns. Campaign creation/loading, character management, combat tracking, NPC generation, dice rolling, an optional live display companion, and full session state — all persisted across sessions. Provides the dnd skill; invoke as /dm:dnd, or just describe what.