What is hermes-cost-audit?

A measurement and review workflow for Hermes model-token spending. It examines prompt sizes, recent usage, models, and settings, then proposes savings without changing anything automatically.

Use it to investigate expensive Hermes usage, compare recent spending, and find measured opportunities to reduce it.

It shows which fixed prompts, tools, and model calls account for cost before you approve a change.

hermes-cost-audit — Skill for Claude Code and Codex by OnlyTerp

Install

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.

Any agent

npx skills add OnlyTerp/hermes-optimization-guide --skill hermes-cost-audit

Clone the repo

git clone --depth 1 https://github.com/OnlyTerp/hermes-optimization-guide

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/onlyterp/hermes-optimization-guide/hermes-cost-audit/github.svg)](https://agentmods.dev/skills/onlyterp/hermes-optimization-guide/hermes-cost-audit)

Your own site

<a href="https://agentmods.dev/skills/onlyterp/hermes-optimization-guide/hermes-cost-audit"><img src="https://agentmods.dev/badge/skills/onlyterp/hermes-optimization-guide/hermes-cost-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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/onlyterp/hermes-optimization-guide/hermes-cost-audit"><img src="https://agentmods.dev/badge/skills/onlyterp/hermes-optimization-guide/hermes-cost-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 14 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 916 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00014	$0.00916
Opus 5.5	$0.00006	$0.00366
Sonnet 5.5	$0.00003	$0.00183
Haiku 4.5	$0.00001	$0.00092

Measured 21d ago against content hash 8d53da367212, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

hermes-cost-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 21d 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.

skills/hermes-cost-audit/SKILL.md · 83 lines

How it starts

The opening of the file, as written. The whole thing — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Hermes Cost Audit

Find where this Hermes install spends tokens and propose the cuts with the best payoff, backed by measurements taken on this machine.

When to Use

The user asks why Hermes is expensive, where tokens go, or how to spend less.
As a weekly cron job that reports spend and new savings opportunities.

Procedure

Run every command with the terminal tool. Read, measure, and report. Do not change any setting until the user approves the specific change.

Measure the fixed prefix that every model call carries, per surface actually in use:

hermes tools --summary                      # which platforms have which toolsets
hermes prompt-size --json                   # CLI
hermes prompt-size --platform telegram --json   # repeat for each platform in use

Record system prompt bytes, skills index bytes, tool schema bytes, and the three largest toolsets per platform.

Measure real spend over time:

hermes insights --days 7
hermes insights --days 30

Read the settings that drive cost. auto on an auxiliary task means the main model:

hermes config get model
hermes config get auxiliary.compression.model
hermes config get auxiliary.title_generation.model
hermes config get auxiliary.approval.model
hermes config get auxiliary.background_review.model
hermes config get delegation.model
hermes config get cron.model
hermes config get agent.disabled_toolsets
hermes config get prompt_caching.cache_ttl

Check for these patterns and estimate the saving of each from the step 1 numbers (roughly 4.2–4.5 bytes per token):

Pattern	Proposal
Auxiliary tasks on `auto` while the main model is expensive	Route compression, titles, approvals, and background review to a cheap model. Leave `auxiliary.vision` unset when the main model can see images: an explicit vision backend makes every image go through a text description.
A toolset enabled on a platform that never uses it (for example `browser` or `tts` on a chat bot)	`hermes tools disable --platform <platform> <toolset>`
Heavy delegation with `delegation.model` unset	A cheaper model for subagents
Several cron jobs with `cron.model` unset	A cheap default for unpinned cron jobs
A large project context file where the user usually works	Shorten it. Move detail into subdirectory files or skills.
Long pauses between turns on a Claude model with `cache_ttl: 5m`	Consider `1h` (cache writes cost 2x, but pauses stop re-reading at full price)

Read the full file on GitHub · 83 lines

Changes

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.

21d ago First seen · 83 lines · 14 tokens per session scan A 8d53da367212

Subscribe to this mod's changes

hermes-cost-audit is a skill published in the GitHub repository OnlyTerp/hermes-optimization-guide (696 stars, last pushed 22d ago), licensed MIT. It adds 14 tokens to every session and 916 once invoked, about $0.0001 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 2026-09-25.

Analytics and metrics

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens