Borrowing it
Nothing to install: this file belongs to pacphi/agentic-kit. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/pacphi/agentic-kit/main/.agents/skills/ak-token-audit/SKILL.mdgit clone --depth 1 https://github.com/pacphi/agentic-kitWrote 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/skills/pacphi/agentic-kit/ak-token-audit)<a href="https://agentmods.dev/skills/pacphi/agentic-kit/ak-token-audit"><img src="https://agentmods.dev/badge/skills/pacphi/agentic-kit/ak-token-audit/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/skills/pacphi/agentic-kit/ak-token-audit"><img src="https://agentmods.dev/badge/skills/pacphi/agentic-kit/ak-token-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00076 | $0.01054 |
| Opus 5.5 | $0.00030 | $0.00422 |
| Sonnet 5.5 | $0.00015 | $0.00211 |
| Haiku 4.5 | $0.00008 | $0.00105 |
Grade A, and why
ak-token-audit 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 23d 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.
How it starts
The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Token audit
A standalone, offline picture of Claude Code usage, built from the local session transcripts in
~/.claude/projects/**/*.jsonl (each assistant message records its tokens, tool calls and model).
It is the maintainer's second opinion beside ak usage and the dashboard's Usage view, and it
answers two questions: where is the usage going, and is any of it runaway automation.
When to use
The maintainer asks where tokens or usage are going, why usage is so high, what a week of Claude Code activity looked like, or says the plan is being burned or limits are being hit.
Steps
-
From the repository root, run the engine (stdlib-only Python 3; if
python3is missing, say so and stop):python3 .claude/skills/ak-token-audit/scripts/ruflo-token-audit.py --days 7Use the window the maintainer gives ("past month" is
--days 30).--top Nwidens each section,--jsonprints machine-readable output and--no-daemonsskips the process list. -
Read the whole report, then lead with the headline. Synthesize; do not echo the sections:
Section Read it for BY MODEL The model mix; a model name alone does not show whether work is interactive or automated SESSIONS PER DAY Volume and bursts to investigate; counts alone do not prove automation ACTIVITY BY HOUR Timing patterns to line up with the maintainer's own activity and with processes TOOL USAGE What the work is (Bash, Read and Edit against Task and MCP calls) MCP USAGE Calls per MCP server; the cost of loaded tool definitions depends on the host SUBAGENT FAN-OUT Task spawns and the sidechain share: how much is delegated BUSIEST SESSIONS A single runaway conversation shows up here by its token total CACHE EFFICIENCY A high cache-read share is normal and cheap; flag it only with huge automated volume STARTUP CONTEXT TAX The fixed per-session cost ( CLAUDE.md, tool and skill manifests) times the session countRUNNING DAEMONS Live ruflo daemon startprocesses mapped to the top-burn projects
What ships with it
1 file beside SKILL.md 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.
- 23d ago First seen · 77 lines · 76 tokens per session scan A 44e089b05bef
ak-token-audit is a skill published in the GitHub repository pacphi/agentic-kit (61 stars, last pushed yesterday), licensed MIT. It adds 76 tokens to every session and 1,054 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-05.
Other skills, from other repositories
typescript-expert
Expert-level TypeScript development with modern tooling, advanced types, and best practices. Use this skill for TypeScript projects requiring type-safe code, modern bundling, and comprehensive testing.
electron-expert
Expert in Electron framework, desktop app development, IPC, and cross-platform packaging. Use when the user mentions desktop, Node.js, cross platform, Windows, macOS, or Linux, or when the task involves Electron Architecture, Process Types, IPC Communication, or App Lifecycle.
nodejs-expert
Expert-level Node.js backend development with Express, async patterns, streams, performance optimization, and production best practices. Use when the user mentions JavaScript, backend, Express, or async, or when the task involves Modern Node.js Features, Express Framework, File System Operations, or HTTP Requests.
nestjs-expert
Expert knowledge in NestJS framework, modular architecture, dependency injection, TypeORM/Prisma persistence, microservices transports, and enterprise-grade Node.js backend development. Use when the user mentions NestJS, Nest CLI, decorators, dependency injection, guards, interceptors, or backend microservices, or…
build-feature
Full pipeline: evaluation -> spec -> implementation -> review -> QA.
council
Convenes multiple agents to debate an important decision.