Prompt Caching

Cross-provider (Anthropic/OpenAI/Bedrock/Gemini) prompt-caching diagnostic playbook — four of its own linked reference files are missing from the repo.

Works with setup

Test report

Verdict
Works with setup
Score
7.2/10
Tested
Jul 14, 2026
Environment
Claude Code 2.x (agent harness)
Upstream re-checked
Aug 10, 2026 · d051749

Nails the actual diagnostic playbook — demands the real hit-rate number, walks the cache-boundary rule, flags Anthropic's 125%-write/10%-read billing asymmetry — but four reference files it points to (references/anthropic-caching.md, openai-caching.md, gemini-caching.md, cost-math.md) simply aren't in the repo; a user has to source those provider deep-dives themselves. (Listed here under our older name; the skill's own name is "Prompt Caching".)

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

  • Installs cleanly 3/5
  • Triggers reliably 5/5
  • Output vs. baseline 8/10
  • Docs & honesty 2/5

What Prompt Caching does

A decision-flow-and-rules skill for diagnosing and fixing prompt-cache misses across Anthropic, OpenAI, Bedrock, and Gemini: when to cache, the 'static above / dynamic below' boundary rule, provider-specific TTL and billing notes, and a hard line that you must measure hit rate before tuning. It fires on cost/latency questions about repeated LLM calls. The SKILL.md itself links out to four deeper reference files (references/anthropic-caching.md, openai-caching.md, gemini-caching.md, cost-math.md) that do not exist anywhere in the repo, so the core guidance works standalone but the promised deep-dives don't.

How to install Prompt Caching

git clone https://github.com/cobusgreyling/agent-skills
cd agent-skills
mkdir -p ~/.claude/skills
cp -r skills/prompt-caching ~/.claude/skills/prompt-cache-optimizer

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Prompt Caching

  • /prompt-cache-optimizer Cross-provider (Anthropic/OpenAI/Bedrock/Gemini) prompt-caching diagnostic playbook — four of its own linked reference files are missing from the repo.

It also activates on plain-language prompts like these:

  • Reorder this prompt so the stable prefix caches better
  • Improve our cache-hit rate on this agent workload
  • Move volatile data to the end of this prompt for caching

Frequently asked questions

Is the Prompt Caching skill free?
Yes. The skill itself is free from cobusgreyling/agent-skills. SkillProof publishes the install command and an independent test verdict at no cost.
Does Prompt Caching work with Claude Code?
We tested it with Claude Code 2.x (agent harness) on Jul 14, 2026. Verdict: Works with setup. Nails the actual diagnostic playbook — demands the real hit-rate number, walks the cache-boundary rule, flags Anthropic's 125%-write/10%-read billing asymmetry — but four reference files it points to (references/anthropic-caching.md, openai-caching.md, gemini-caching.md, cost-math.md) simply aren't in the repo; a user has to source those provider deep-dives themselves. (Listed here under our older name; the skill's own name is "Prompt Caching".)
What is the Prompt Caching SkillProof Score?
7.2/10 — installs cleanly 3/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 2/5.
How do I install Prompt Caching?
Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Prompt Caching with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.