s4h-information-compression

s4h-information-compression is a skill for Claude Code from human-avatar/skills-for-humanity. It costs 71 tokens per session (1,775 once invoked), scanned A, original, MIT.

A way to shorten reports, explanations, or models while deciding which information must be preserved. It distinguishes lossless compression, which keeps everything, from lossy compression, which removes some detail.

In plain words
What is it for?
Use it to reduce a document or decision brief, identify essential content, and check whether a summary is still accurate.
Why use it?
It helps cut padding without losing the important point, and makes the trade-off clear when shortening requires discarding information.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: names the AskUserQuestion tool.

Part of the skills-for-humanity plugin — 197 skills, 1 hook shipped together

Good fit Use it to reduce a document or decision brief, identify essential content, and check whether a summary is still accurate.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/human-avatar/skills-for-humanity/s4h-information-compression
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 human-avatar/skills-for-humanity --skill s4h-information-compression
Clone the repo
git clone --depth 1 https://github.com/human-avatar/skills-for-humanity

Made for: Claude Code.

Or install skills-for-humanity, the plugin that ships this one along with the rest of its 197 skills, 1 hook.

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.

agentmods badge for s4h-information-compression

README.md
[![agentmods](https://agentmods.dev/badge/skills/human-avatar/skills-for-humanity/s4h-information-compression/github.svg)](https://agentmods.dev/skills/human-avatar/skills-for-humanity/s4h-information-compression)
Your own site
<a href="https://agentmods.dev/skills/human-avatar/skills-for-humanity/s4h-information-compression"><img src="https://agentmods.dev/badge/skills/human-avatar/skills-for-humanity/s4h-information-compression/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.

agentmods 80×15 button for s4h-information-compression

Your own site · 80×15
<a href="https://agentmods.dev/skills/human-avatar/skills-for-humanity/s4h-information-compression"><img src="https://agentmods.dev/badge/skills/human-avatar/skills-for-humanity/s4h-information-compression.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 71 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,775 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. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00071 $0.01775
Opus 5 $0.00036 $0.00888
Sonnet 5 $0.00014 $0.00355
Haiku 4.5 $0.00007 $0.00178

Measured 9d ago against content hash c7d45e36a481, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

s4h-information-compression 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 9d 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/s4h-information-compression/SKILL.md · 118 lines

How it starts

The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Information: Compression

Every representation is a compression of reality. A report compresses events. A model compresses a domain. An explanation compresses understanding. A decision brief compresses weeks of analysis. The question is never whether to compress — it's whether the compression preserves what matters.

Claude Shannon proved that there is a theoretical lower bound on how far any lossless compression can go: you cannot compress below the entropy of the source without discarding information. This is the source coding theorem. Applied practically: there is always a floor. A message with genuine information content cannot be shortened indefinitely without loss — and often the thing being shortened to below-floor is not actually long; it's precise. The problem is usually the reverse: sources that are padded, redundant, or poorly structured can be dramatically shortened without any loss, because they weren't near the floor to begin with.

The critical distinction is lossy vs. lossless. Lossless compression preserves every bit of the original — the compression is reversible. Lossy compression discards some information permanently, in exchange for a representation that is smaller or more usable. Both are legitimate; the choice depends entirely on what the information is for, who will use it, and what the cost of loss is. An executive summary is lossy by design. A legal contract is lossless by necessity. The skill is making the trade-off explicit — and knowing what you are throwing away.

James Gleick's The Information notes that compression is cognition: the brain is fundamentally a compression engine. Understanding is compression. This is why the best explanations are short: not because they're simpler, but because they've found the structure that allows reconstruction from minimal representation. Teaching someone a principle rather than a list of cases is compression.


Your Process

Step 1: Identify What the Representation Is For Before deciding what to cut, establish the purpose of the compressed form. Who is it for? What decision or action does it need to enable? What understanding must be present in the receiver after they've processed the compressed version? The appropriate compression strategy is entirely downstream of this answer.

Read the full file on GitHub · 118 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.

  1. 9d ago First seen · 118 lines · 71 tokens per session scan A c7d45e36a481

Subscribe to this mod's changes

s4h-information-compression is a skill published in the GitHub repository human-avatar/skills-for-humanity (223 stars, last pushed 1mo ago), licensed MIT. It adds 71 tokens to every session and 1,775 once invoked, about $0.0004 per session on Opus 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 2026-09-03.

Related

Other skills, from other repositories

skill-deck

Generate slide deck presentations from briefs — use when you need slides, pitch decks, or visual summaries.

nyldn/claude-octopus · 23 tokens

add-onedrive

Adds OneDrive for Business connector to a Power Apps code app. Use when uploading, downloading, listing, or managing files in OneDrive.

microsoft/power-platform-skills · 33 tokens

resume

Generate a tailored resume AND cover letter for a job description via the native four-role Resume Team, score both against ATS and HR rubrics, create DOCX files, and update the tracker. Use when the user pastes a job description and wants a complete application package (resume plus cover letter) with dual scoring and…

jananthan30/Resume-Builder · 72 tokens

codex-review-doc

Review documents using Codex exec. Supports review loop with context preservation.

sd0xdev/sd0x-harness · 19 tokens

migrate

Scan the project and migrate existing documentation into OKF format. Use when the user wants to populate the knowledge base from existing docs, README, schemas, API specs, runbooks, or any structured project knowledge. Trigger on "migrate to okf", "populate the knowledge base", "scan my docs", "convert docs to okf"…

guhcostan/claude-mega-brain · 147 tokens

fulcra-lab-results

Parse a lab-report PDF (LabCorp, Quest, hospital labs — any provider) into one canonical Fulcra data track per marker (LDL-C, HbA1c, TSH, ferritin, …). You read the PDF; the fulcra-labs engine normalizes units, validates, and ingests. Verify-before-ingest: nothing is written without passing validation or explicit…

ashfulcra/fulcra-tools · 114 tokens