Agent skill

Aider One-Hour Cache TTL

by OnlyTerp in OnlyTerp/prompt-cache-skills

Patches Aider to request Anthropic's one-hour prompt cache lifetime, so long pauses between turns no longer expire the cache or need paid keepalive pings.

Custom licenceAuto-check passedAI & LLM Engineering

Install Aider One-Hour Cache TTL

skills CLI
$ npx skills add OnlyTerp/prompt-cache-skills --skill aider-1h-ttl -a claude-code

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

GitHub CLI
$ gh skill install OnlyTerp/prompt-cache-skills aider-1h-ttl --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/OnlyTerp/prompt-cache-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/aider-1h-ttl .claude/skills/aider-1h-ttl && 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/

Facts

Skill name
aider-1h-ttl
GitHub stars
114
Token cost
~990 tokens
SKILL.md length
244 words
Files
1
Skills in repo
13
Repo updated
First seen
Licence
Custom licence

At a glance

Patches Aider to request Anthropic's one-hour prompt cache lifetime, so long pauses between turns no longer expire the cache or need paid keepalive pings.

  • Works in 6 steps: Run with --cache-prompts… → Capture wire. → Request headers should include → …
  • Aider sessions with long pauses between prompts
  • SKILL.md covers Target, Symptom, Fix (three parts) and Verify, plus 1 more section
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Aider caches prompts for five minutes by default, so a pause of ten minutes to read code and think lets the cache lapse between turns. The usual workaround, `--cache-keepalive-pings`, sends a trivial request every five minutes and you pay for each one. This skill wires in Anthropic's extended-cache-ttl beta instead, which raises the lifetime to one hour at a higher write cost, about 2x instead of 1.25x, with a break-even near four reads.

The change has three parts in Aider's source: add an extended cache TTL flag in `aider/args.py`, pass it through the message builder in `aider/coders/chat_chunks.py`, and set the `anthropic-beta` header in `aider/sendchat.py` when it is on. To verify, run with `--cache-prompts --extended-cache-ttl --no-stream`, confirm the header and a `ttl` of `1h` on the cache control blocks, wait six minutes, send another prompt and check that cache-read tokens still cover the prefix.

When your agent uses it

  • Aider sessions with long pauses between prompts
  • Cutting the cost of keepalive pings in Aider
  • Checking that an Aider cache still holds after more than five minutes

Example prompts

  • “Patch my Aider checkout to support the one-hour cache TTL beta.”
  • “Add an --extended-cache-ttl flag to Aider and check the beta header on the wire.”
  • “Verify that Aider's cache survives a six-minute gap after the TTL change.”

Requirements

  • A local checkout of the Aider source
  • Aider configured with an Anthropic model

Workflow steps

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

  1. Run with --cache-prompts --extended-cache-ttl --no-stream.
  2. Capture wire.
  3. Request headers should include
  4. Request body cache_control blocks should be
  5. Wait 6 minutes. Send another prompt.
  6. Response usage.cache_read_input_tokens should still cover the

What it can do on your machine

Read from SKILL.md and the folder at commit 5b58ae2. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    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 diff).

    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

Aider One-Hour Cache TTL loads about 990 tokens when it runs. Until then it costs about 32 tokens; SKILL.md has 244 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~32
When it runs · the whole SKILL.md, loaded when a task matches
~990

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 passed

The automated check found no risky patterns in SKILL.md.

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

Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 244 words (~990 tokens).

name
aider-1h-ttl
target_harness
Aider
target_repo
Aider-AI/aider
target_files
aider/coders/chat_chunks.py, aider/sendchat.py, aider/args.py
target_commit
3ec8ec5a
estimated_savings
Eliminates --cache-keepalive-pings billing waste; preserves cache across multi-hour thinking pauses

Read the full SKILL.md on GitHub

Files

Just SKILL.md in skills/aider-1h-ttl of OnlyTerp/prompt-cache-skills.

Open the folder on GitHubat commit 5b58ae2

Compare with similar skills

Aider One-Hour Cache TTL 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.

Aider One-Hour Cache TTL compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Aider One-Hour Cache TTL this skillOnlyTerp/prompt-cache-skills114—~990Automated safety check: PassCustom licence
Claude APIloulanyue/awesome-claude-notes2722 repos~2.1kAutomated safety check: PassMIT
Claude API Developmentwarpdotdev/warp65k3 repos~8.2kAutomated safety check: PassApache-2.0
ClawRouter LLM GatewayBlockRunAI/ClawRouter6.6k—~6.8kAutomated safety check: PassMIT
AIbutterbase-ai/butterbase-skills534—~1.1kAutomated safety check: PassMIT
LLM Cost Optimizationsickn33/agentic-awesome-skills47k1 repos~2.5kAutomated safety check: PassMIT

Similar skills

  • Claude API

    loulanyue/awesome-claude-notes

    Anthropic Claude API patterns for Python and TypeScript. An agent skill from loulanyue/awesome-claude-notes.

    272 GitHub starsUsed in 2 repos~2.1k tokens
    AI & LLM EngineeringAuto-check passed
  • Claude API Development

    warpdotdev/warp

    Guides building, debugging and tuning apps on the Claude API and Anthropic SDK, including prompt caching, and migrating code between Claude model versions.

    65k GitHub starsUsed in 3 repos~8.2k tokens
    AI & LLM EngineeringAuto-check passed
  • ClawRouter LLM Gateway

    BlockRunAI/ClawRouter

    Describes ClawRouter, a local proxy that forwards each LLM request to the blockrun.ai gateway, which routes to a cheaper capable model, paid by USDC wallet or API key credit.

    6.6k GitHub stars~6.8k tokensUpdated 5 days ago
    AI & LLM EngineeringAuto-check passed
  • AI

    butterbase-ai/butterbase-skills

    A skill your agent uses when calling the app's AI gateway from agent tools — chat completions, embeddings, listing models, configuring defaults or BYOK, reading token/cost usage

    534 GitHub stars~1.1k tokensUpdated 5 days ago
    AI & LLM EngineeringAuto-check passed
  • LLM Cost Optimization

    sickn33/agentic-awesome-skills

    Reduce LLM API and infrastructure costs through model selection, prompt caching, batching, caching, quantization, and self-hosting strategies.

    47k GitHub starsUsed in 1 repo~2.5k tokens
    AI & LLM EngineeringAuto-check passed
  • Cost Aware LLM Pipeline

    aAAaqwq/AGI-Super-Team

    Cost optimization patterns for LLM API usage — model routing by task complexity, budget tracking, retry logic, and prompt caching.

    105 GitHub starsUsed in 5 repos~1.4k tokens
    AI & LLM EngineeringAuto-check passed

More from OnlyTerp/prompt-cache-skills

All 13 skills in this repo
  • Aider Prompt Caching Default

    OnlyTerp/prompt-cache-skills

    Patch recipe that turns on Aider's --cache-prompts by default for models that support caching, with a verification procedure against the wire request.

    114 GitHub stars~638 tokensUpdated 1 mo ago
    Auto-check passed
  • Cline Prompt Cache Ladder Check

    OnlyTerp/prompt-cache-skills

    Verifies how Cline marks Anthropic prompt-cache breakpoints on the wire, then adds a tools breakpoint and tunes the cache TTL in its AI SDK code.

    114 GitHub stars~2k tokensUpdated 1 mo ago
    Auto-check passed
  • Cline OpenAI Prompt Cache Key Fix

    OnlyTerp/prompt-cache-skills

    Patch guide for adding a stable per-task prompt_cache_key to Cline's OpenAI native provider so cached token counts stop reading as zero.

    114 GitHub stars~868 tokensUpdated 1 mo ago
    Auto-check passed
  • Cline Pin Timestamp

    OnlyTerp/prompt-cache-skills

    Cline's system prompt includes a timestamp that may be recomputed per request, invalidating the system-prompt cache.

    114 GitHub stars~751 tokensUpdated 1 mo ago
    Auto-check passed
  • Continue Enable Defaults

    OnlyTerp/prompt-cache-skills

    Continue's prompt caching is opt-in via config and off by default.

    114 GitHub stars~977 tokensUpdated 1 mo ago
    Auto-check passed
  • Continue Gemini Explicit

    OnlyTerp/prompt-cache-skills

    Continue's Gemini provider doesn't use the cachedContents API at all.

    114 GitHub stars~1.1k tokensUpdated 1 mo ago
    Auto-check passed

Questions about Aider One-Hour Cache TTL

What does Aider One-Hour Cache TTL do?

Patches Aider to request Anthropic's one-hour prompt cache lifetime, so long pauses between turns no longer expire the cache or need paid keepalive pings. Aider caches prompts for five minutes by default, so a pause of ten minutes to read code and think lets the cache lapse between turns. The usual workaround, `--cache-keepalive-pings`, sends a trivial request every five minutes and you pay for each one.

When should I use Aider One-Hour Cache TTL?

Aider One-Hour Cache TTL fits situations like: aider sessions with long pauses between prompts; cutting the cost of keepalive pings in Aider; checking that an Aider cache still holds after more than five minutes.

How do I install Aider One-Hour Cache TTL in Claude Code?

Run `npx skills add OnlyTerp/prompt-cache-skills --skill aider-1h-ttl -a claude-code`. Or copy the skill folder (skills/aider-1h-ttl in OnlyTerp/prompt-cache-skills) into .claude/skills/aider-1h-ttl in your project. Claude Code loads it when a task matches its description.

How do I install Aider One-Hour Cache TTL in Codex?

Run `npx skills add OnlyTerp/prompt-cache-skills --skill aider-1h-ttl -a codex`. Or copy the skill folder (skills/aider-1h-ttl in OnlyTerp/prompt-cache-skills) into .agents/skills/aider-1h-ttl in your project. Codex loads it when a task matches its description.

Can I use Aider One-Hour Cache TTL 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 OnlyTerp/prompt-cache-skills --skill aider-1h-ttl -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/aider-1h-ttl, .gemini/skills/aider-1h-ttl, .github/skills/aider-1h-ttl and .opencode/skills/aider-1h-ttl in your project.

What does Aider One-Hour Cache TTL need to run?

SKILL.md names no scripts, command-line tools or credentials: Aider One-Hour Cache TTL is instructions for the agent only. Our summary lists: A local checkout of the Aider source; Aider configured with an Anthropic model.

Does Aider One-Hour Cache TTL 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 Aider One-Hour Cache TTL safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Aider One-Hour Cache TTL use?

Aider One-Hour Cache TTL has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.

How many tokens does Aider One-Hour Cache TTL use?

About 990 tokens (SKILL.md is roughly 4k 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 Aider One-Hour Cache TTL?

Skills that share tags, products or a category with Aider One-Hour Cache TTL: Claude API (loulanyue/awesome-claude-notes, 272 stars), Claude API Development (warpdotdev/warp, 65k stars), ClawRouter LLM Gateway (BlockRunAI/ClawRouter, 6.6k stars) and AI (butterbase-ai/butterbase-skills, 534 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Aider One-Hour Cache TTL?

OnlyTerp (a GitHub user) maintains it in OnlyTerp/prompt-cache-skills, which has 114 GitHub stars. The repository holds 13 skills in this directory. The repository was last updated on August 29, 2026.

Source: OnlyTerp/prompt-cache-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.