Agent skill

Cache Policy Hit Rate Comparison

Compares cache eviction policies by hit rate across several cache sizes on a trace file using the Caffeine simulator, with CSV tables and a PNG chart.

Apache-2.0Auto-check: notesDevelopment

Install Cache Policy Hit Rate Comparison

skills CLI

$ npx skills add ben-manes/caffeine --skill sim-compare -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install ben-manes/caffeine sim-compare --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/ben-manes/caffeine.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/sim-compare .claude/skills/sim-compare && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "sim-compare" agent skill from https://github.com/ben-manes/caffeine/tree/master/.claude/skills/sim-compare into .claude/skills/sim-compare/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-compare", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/ben-manes/caffeine/tree/master/.claude/skills/sim-compare

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add ben-manes/caffeine --skill sim-compare -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install ben-manes/caffeine sim-compare --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/ben-manes/caffeine.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/sim-compare .agents/skills/sim-compare && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "sim-compare" agent skill from https://github.com/ben-manes/caffeine/tree/master/.claude/skills/sim-compare into .agents/skills/sim-compare/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-compare", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add ben-manes/caffeine --skill sim-compare -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install ben-manes/caffeine sim-compare --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/ben-manes/caffeine.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/sim-compare .cursor/skills/sim-compare && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "sim-compare" agent skill from https://github.com/ben-manes/caffeine/tree/master/.claude/skills/sim-compare into .cursor/skills/sim-compare/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-compare", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/ben-manes/caffeine.git --path .claude/skills/sim-compare

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add ben-manes/caffeine --skill sim-compare -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install ben-manes/caffeine sim-compare --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/ben-manes/caffeine.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/sim-compare .gemini/skills/sim-compare && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "sim-compare" agent skill from https://github.com/ben-manes/caffeine/tree/master/.claude/skills/sim-compare into .gemini/skills/sim-compare/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-compare", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install ben-manes/caffeine sim-compare

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add ben-manes/caffeine --skill sim-compare -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/ben-manes/caffeine.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/sim-compare .github/skills/sim-compare && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "sim-compare" agent skill from https://github.com/ben-manes/caffeine/tree/master/.claude/skills/sim-compare into .github/skills/sim-compare/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-compare", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add ben-manes/caffeine --skill sim-compare -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install ben-manes/caffeine sim-compare --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/ben-manes/caffeine.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/sim-compare .opencode/skills/sim-compare && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "sim-compare" agent skill from https://github.com/ben-manes/caffeine/tree/master/.claude/skills/sim-compare into .opencode/skills/sim-compare/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sim-compare", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: sim-compare
GitHub stars: 18k
Token cost: ~939 tokens
SKILL.md length: 368 words
Files: 1
Skills in repo: 33
Repo updated: Oct 6, 2026
First seen: Oct 7, 2026
Licence: Apache-2.0

At a glance

Compares cache eviction policies by hit rate across several cache sizes on a trace file using the Caffeine simulator, with CSV tables and a PNG chart.

Works in 7 steps: Identify the trace format. Check the… → Select policies to compare. Policy names… → Choose cache sizes. Use a geometric… → …
Comparing Caffeine against other eviction policies on a real trace
SKILL.md covers Input and Workflow
Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

This skill runs a hit-rate comparison of cache policies in the Caffeine simulator for a trace file you provide. It first identifies the trace format from the file extension and contents and from the simulator's reference configuration, using the `format:path` syntax such as `lirs:trace.gz`. It then picks policies named in `category.PolicyName` form, with hyphenated categories, always including the production Caffeine policy, the theoretical optimal and unbounded baselines, LRU, Window TinyLFU and its hill-climbing variant.

Extra competitors are added to suit the trace: recency-heavy, frequency-heavy, scan-resistant or size-aware policies. Cache sizes follow a geometric progression from small values up to near the working set, typically five to eight of them. The run uses the Gradle `simulator:simulate` task with a maximum-size list and the Hit Rate metric, with system properties to override the trace and policies, or `simulator:run` for a single size. Output is one CSV per size, a combined CSV with policies as rows and sizes as columns, and a PNG line chart.

When your agent uses it

Comparing Caffeine against other eviction policies on a real trace
Choosing which cache sizes to simulate for a given workload
Producing hit-rate charts and CSV tables to support a cache tuning decision

Example prompts

“Compare policies on the lirs trace in ./traces/ds1.gz across several cache sizes.”
“Which eviction policies belong in the comparison for a scan-heavy trace, and which sizes?”
“Run the simulator on this trace at a single cache size of 5000 and report the hit rates.”

Requirements

A checkout of the Caffeine repository with its Gradle wrapper
A cache trace file
Pre-approved tools (allowed-tools): Read, Grep, Glob, Bash

Workflow steps

7 steps, taken from the first numbered list in SKILL.md.

Identify the trace format. Check the file extension and contents
Select policies to compare. Policy names use category.PolicyName format.
Choose cache sizes. Use a geometric progression covering the working set
Run the simulation. Use the Gradle task
Read and interpret results. The simulate task produces
Explain findings. For each notable result
Report. Present

What it can do on your machine

Read from SKILL.md and the folder at commit e972fb0. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves these tools, so the agent can use them without asking each time:
- Read
- Grep
- Glob
- Bash
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md (its code samples are bash).
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Cache Policy Hit Rate Comparison loads about 939 tokens when it runs. Until then it costs about 22 tokens; SKILL.md has 368 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~22
When it runs · the whole SKILL.md, loaded when a task matches: ~939

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
```
allowed-tools: Read, Grep, Glob, Bash
```

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from ben-manes/caffeine at commit e972fb0, republished under its Apache-2.0 licence (© ben-manes). 368 words, ~939 tokens.

Download SKILL.mdSave it as .claude/skills/sim-compare/SKILL.md (or your agent's skills folder).

name: sim-compare
description: Run cache policy hit rate comparison across multiple cache sizes with charts
allowed-tools: Read, Grep, Glob, Bash
argument-hint: <trace-file> [policies...] [sizes...]
context: fork
disable-model-invocation: true

Run a comprehensive cache policy comparison for the given trace.

Input

Trace file: $ARGUMENTS

If no policies or sizes are specified, use sensible defaults based on the trace.

Workflow

Identify the trace format. Check the file extension and contents:
- .gz files: try common formats (lirs, arc, etc.)
- Look at simulator/src/main/resources/reference.conf for format options
- Use format:path syntax (e.g., lirs:trace.gz)
Select policies to compare. Policy names use category.PolicyName format. Note: config categories use hyphens (two-queue, greedy-dual), not underscores. Each policy is paired with configured admission filters (default: Always, TinyLfu, Clairvoyant), creating multiple instances per policy name.

Include at minimum:
- product.Caffeine (the production implementation)
- opt.Clairvoyant (theoretical optimal, upper bound)
- opt.Unbounded (infinite cache, ceiling)
- linked.Lru (baseline)
- sketch.WindowTinyLfu (research W-TinyLFU)
- sketch.HillClimberWindowTinyLfu (adaptive variant)
- Add relevant competitors based on trace characteristics:
  - For recency-heavy: linked.S4Lru, adaptive.Arc
  - For frequency-heavy: linked.Lfu, irr.Lirs
  - For scan-resistant: two-queue.TwoQueue, two-queue.S3Fifo
  - For size-aware traces: greedy-dual.Gdsf, greedy-dual.Camp
Choose cache sizes. Use a geometric progression covering the working set:
- Start small (e.g., 100), end near working set size
- 5-8 sizes: e.g., 100,500,1_000,2_500,5_000,10_000,25_000
- If the trace has few distinct keys, reduce the range

Run the simulation. Use the Gradle task:

bash

./gradlew simulator:simulate -q \
  --maximumSize=100,500,1000,2500,5000,10000 \
  --metric="Hit Rate" \
  --title="Description" \
  --theme=light \
  --outputDir=build/reports/sim

Override the trace and policies via system properties appended to the command:

bash

-Dcaffeine.simulator.files.paths.0="format:path/to/trace"
-Dcaffeine.simulator.policies.0=product.Caffeine
-Dcaffeine.simulator.policies.1=opt.Clairvoyant
# ... etc

Note: for single-size runs, use ./gradlew simulator:run -q with -Dcaffeine.simulator.maximum-size=N instead of simulator:simulate.

Read and interpret results. The simulate task produces:
- Individual CSV per cache size
- Combined CSV (policies as rows, sizes as columns)
- PNG chart (line graph of metric vs cache size) Read the CSV output files in the output directory:
- Compare hit rates across policies at each cache size
- Identify the crossover points where one policy overtakes another
- Note the gap between Caffeine and Clairvoyant (theoretical ceiling)
Explain findings. For each notable result:
- WHY does policy X beat policy Y on this trace?
- What trace characteristic drives the difference? (frequency bias, recency bias, scan patterns, temporal shifts)
- How close is Caffeine to optimal? Where does it lose?
- Reference the relevant research paper if applicable:
  - TinyLFU paper for admission filter behavior
  - Adaptive paper for hill climber effectiveness
  - See .claude/docs/research-foundations.md for paper-to-code mapping
Report. Present:
- Summary table of hit rates at each cache size
- Key takeaways (2-3 sentences)
- Notable policy behaviors
- Path to generated chart PNG

Files

Just SKILL.md in .claude/skills/sim-compare of ben-manes/caffeine.

Open the folder on GitHubat commit e972fb0

Compare with similar skills

Cache Policy Hit Rate Comparison next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Cache Policy Hit Rate Comparison compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Cache Policy Hit Rate Comparison this skillben-manes/caffeine	18k	—	~939	Automated safety check: Notes	Apache-2.0	yesterday
Groovy 5 Developer Guideapache/grails-core	2.9k	—	~3k	Automated safety check: Pass	Apache-2.0	today
Keybase RPC Log Analysiskeybase/client	9.3k	—	~3k	Automated safety check: Pass	BSD-3-Clause	today
Performance CheckZeroDeng01/sublinkPro	1.7k	—	~1.8k	Automated safety check: Pass	MIT	2 days ago
Minecraft CI ReleaseJahrome907/minecraft-agent-skills	166	—	~3.6k	Automated safety check: Pass	MIT	24 days ago
Releasesol4k/sol4k	135	—	~949	Automated safety check: Pass	Apache-2.0	9 days ago

Similar skills

Groovy 5 Developer Guide
apache/grails-core
Guidance for Groovy 5 work in Grails projects: syntax, closures, traits, DSLs, metaprogramming, Spock tests, static compilation and Java 21 integration.
2.9k GitHub stars~3k tokensUpdated today
DevelopmentAuto-check passed
Keybase RPC Log Analysis
keybase/client
Captures a clean Keybase service log and analyzes it for redundant, duplicated or looping RPCs, then checks whether a caching fix reduced the calls.
9.3k GitHub stars~3k tokensUpdated today
DevelopmentAuto-check passed
Performance Check
ZeroDeng01/sublinkPro
Checklist for reviewing code changes that touch queries, APIs, rendering, caching or algorithms for performance, scalability and resource-usage problems.
1.7k GitHub stars~1.8k tokensUpdated 2 days ago
DevelopmentAuto-check passed
Minecraft CI Release
Jahrome907/minecraft-agent-skills
Set up and review CI, artifact publishing, versioning, and release management for Minecraft 26.x or legacy 1.21.x mods and Paper plugins.
166 GitHub stars~3.6k tokensUpdated 24 days ago
DevelopmentAuto-check passed
Release
sol4k/sol4k
Bump the sol4k library version everywhere, open a release PR, and draft GitHub release notes.
135 GitHub stars~949 tokensUpdated 9 days ago
DevelopmentAuto-check passed
Java 21 Developer Guide for Grails
apache/grails-core
Guide for writing modern Java 21 in a Grails and Groovy codebase: records, sealed classes, pattern matching, text blocks and how Java works alongside Groovy.
2.9k GitHub stars~1.9k tokensUpdated today
DevelopmentAuto-check passed

Questions about Cache Policy Hit Rate Comparison

What does Cache Policy Hit Rate Comparison do?

Compares cache eviction policies by hit rate across several cache sizes on a trace file using the Caffeine simulator, with CSV tables and a PNG chart. This skill runs a hit-rate comparison of cache policies in the Caffeine simulator for a trace file you provide.gz`.

When should I use Cache Policy Hit Rate Comparison?

Cache Policy Hit Rate Comparison fits situations like: comparing Caffeine against other eviction policies on a real trace; choosing which cache sizes to simulate for a given workload; producing hit-rate charts and CSV tables to support a cache tuning decision.

How do I install Cache Policy Hit Rate Comparison in Claude Code?

Run `npx skills add ben-manes/caffeine --skill sim-compare -a claude-code`. Or copy the skill folder (.claude/skills/sim-compare in ben-manes/caffeine) into .claude/skills/sim-compare in your project. Claude Code loads it when a task matches its description.

How do I install Cache Policy Hit Rate Comparison in Codex?

Run `npx skills add ben-manes/caffeine --skill sim-compare -a codex`. Or copy the skill folder (.claude/skills/sim-compare in ben-manes/caffeine) into .agents/skills/sim-compare in your project. Codex loads it when a task matches its description.

Can I use Cache Policy Hit Rate Comparison in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ben-manes/caffeine --skill sim-compare -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sim-compare, .gemini/skills/sim-compare, .github/skills/sim-compare and .opencode/skills/sim-compare in your project.

What does Cache Policy Hit Rate Comparison need to run?

SKILL.md names no scripts, command-line tools or credentials: Cache Policy Hit Rate Comparison is instructions for the agent only. Our summary lists: A checkout of the Caffeine repository with its Gradle wrapper; A cache trace file. Its frontmatter pre-approves these tools: Read, Grep, Glob, Bash.

Does Cache Policy Hit Rate Comparison access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Cache Policy Hit Rate Comparison safe to install?

Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Cache Policy Hit Rate Comparison use?

Cache Policy Hit Rate Comparison is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Cache Policy Hit Rate Comparison use?

About 939 tokens (SKILL.md is roughly 3.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Cache Policy Hit Rate Comparison?

Skills that share tags, products or a category with Cache Policy Hit Rate Comparison: Groovy 5 Developer Guide (apache/grails-core, 2.9k stars), Keybase RPC Log Analysis (keybase/client, 9.3k stars), Performance Check (ZeroDeng01/sublinkPro, 1.7k stars) and Minecraft CI Release (Jahrome907/minecraft-agent-skills, 166 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Cache Policy Hit Rate Comparison?

ben-manes (a GitHub user) maintains it in ben-manes/caffeine, which has 17,880 GitHub stars. The repository holds 33 skills in this directory. The repository was last updated on October 6, 2026.

Source: ben-manes/caffeine on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Cache Policy Hit Rate Comparison

Install Cache Policy Hit Rate Comparison

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Input

Workflow

Files

Compare with similar skills

Similar skills

Groovy 5 Developer Guide

Keybase RPC Log Analysis

Performance Check

Minecraft CI Release

Release

Java 21 Developer Guide for Grails

More from ben-manes/caffeine

Caffeine Cache Optimization Experiments

Git History Bug Audit

Adversarial Codebase Audit

Caffeine Performance Audit

Audit Sibling Divergence

Climber Step Minimization

Questions about Cache Policy Hit Rate Comparison

Install Cache Policy Hit Rate Comparison

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Input

Workflow

Files

Compare with similar skills

Similar skills

More from ben-manes/caffeine

Related

Questions about Cache Policy Hit Rate Comparison