New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Cache

by jeremylongshore in jeremylongshore/tons-of-skills-marketplace

A skill your agent uses when the user wants to view, search, add, edit, prune, archive, or clear hyperflow memory entries.

MITAuto-check passed

Install Cache

skills CLI
$ npx skills add jeremylongshore/tons-of-skills-marketplace --skill cache -a claude-code

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

GitHub CLI
$ gh skill install jeremylongshore/tons-of-skills-marketplace cache --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/jeremylongshore/tons-of-skills-marketplace.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/ai-agency/hyperflow/skills/cache .claude/skills/cache && 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
cache
GitHub stars
2.8k
Token cost
~2.5k tokens
SKILL.md length
941 words
Files
5 (incl. references)
Skills in repo
3,342
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when the user wants to view, search, add, edit, prune, archive, or clear hyperflow memory entries.

  • Works in 8 steps: The compact subcommand handler reads the… → The Date/tag parser splits entries into… → The Compaction Writer is dispatched in a… → …
  • The user wants to view
  • SKILL.md covers Storage, Subcommands, Subcommand Details and Flow, plus 7 more sections
  • Calls python3

What it does

Cache is an agent skill from jeremylongshore/tons-of-skills-marketplace. Use when the user wants to view, search, add, edit, prune, archive, or clear hyperflow memory entries. CRUD interface for .hyperflow/memory/ — never modifies source code, only memory files. Trigger with /hyperflow:cache, "show memory", "search memory for X", "clear memory", "what does hyperflow remember about Y".

Its SKILL.md is about 2.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including reference files (for example `references/DOCTRINE.md`, `references/compaction.md` and `references/memory-system.md`). Compatibility notes: Designed for Claude Code

The repository describes itself as: Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com. The licence is MIT.

When your agent uses it

  • The user wants to view
  • Clear hyperflow memory entries
  • With /hyperflow:cache
  • Search memory for X

Example prompts

  • “show memory”
  • “search memory for X”
  • “clear memory”
  • “/cache”

Requirements

  • Python 3
  • Compatibility (from SKILL.md): Designed for Claude Code
  • Pre-approved tools (allowed-tools): Read, Write, Edit, Bash(ls:*), Bash(mv:*), Bash(rm:*), Glob, Grep, Agent, AskUserQuestion

Workflow steps

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

  1. The compact subcommand handler reads the target memory file (default: learnings.md; pass a path to target another).
  2. The Date/tag parser splits entries into hot (≤7 days, preserved) and eligible (>7 days). Both [domain, type] and legacy backticked…
  3. The Compaction Writer is dispatched in a single batch with all eligible entries.
  4. The Stub formatter renders each replacement line as ### [YYYY-MM-DD] Short title [domain, type] — summarized, see archive/YYYY-MM.md.
  5. The Dedup Reviewer performs source-side stub-line match and archive-side header match (date + title + tags on both sides) to prevent…
  6. The Archive-sidecar writer appends accepted entries to archive/YYYY-MM.md, grouped by each entry's calendar month.
  7. The source file is rewritten with stubs replacing the original entries.
  8. The compact subcommand handler refreshes .hyperflow/memory/.checksums (a memory-scoped sidecar — distinct from .hyperflow/.checksums which…

What it can do on your machine

Read from SKILL.md and the folder at commit cfae287. 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
    • Write
    • Edit
    • Bash(ls:*)
    • Bash(mv:*)
    • Bash(rm:*)
    • Glob
    • Grep
    • Agent
    • AskUserQuestion

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • python3

    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.

  • Compatibility

    Designed for Claude Code

    From compatibility in the SKILL.md frontmatter.

Context cost

Cache loads about 2.5k tokens when it runs, and up to ~20k if it reads all its reference files. Until then it costs about 81 tokens; SKILL.md has 941 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~81
When it runs · the whole SKILL.md, loaded when a task matches
~2.5k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~20k

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

The full file from jeremylongshore/tons-of-skills-marketplace at commit cfae287, republished under its MIT licence (© jeremylongshore). 941 words, ~2,456 tokens.

Download SKILL.mdSave it as .claude/skills/cache/SKILL.md (or your agent's skills folder). This skill also uses 4 other files; get the full folder from GitHub.
name
cache
description
Use when the user wants to view, search, add, edit, prune, archive, or clear hyperflow memory entries. CRUD interface for `.hyperflow/memory/` — never modifies source code, only memory files. Trigger with /hyperflow:cache, "show memory", "search memory for X", "clear memory", "what does hyperflow remember about Y".
allowed-tools
Read, Write, Edit, Bash(ls:*), Bash(mv:*), Bash(rm:*), Glob, Grep, Agent, AskUserQuestion
compatibility
Designed for Claude Code
argument-hint
<show|search|add|edit|prune|archive|clear|stats|migrate|off|compact> [args]
version
3.1.3
license
MIT
tags
memory, persistence, project-state

Cache

CRUD interface for .hyperflow/memory/. Full protocol: memory-system.md.

Storage

All operations target .hyperflow/memory/ at the project root. Never modify source code files — if asked to "remember X about file Y", add a memory entry only, never edit Y.

Subcommands

SubcommandDescription
show [tag]Print index or filter entries by tag
search <query>Full-text search across all memory files
add <category> <title>Append a new entry (prompts for details)
edit <entry-id>Find entry by date+title slug and update in place
pruneRemove stale, superseded, and orphaned entries
archiveMove entries older than 30 days to cold storage
clearWipe all memory (with confirmation, recoverable)
statsCounts, tier breakdown, tag frequency, oldest/newest
migrateImport entries from legacy ~/.claude/hyperflow-memory.md
offDisable memory writes for this session
compactSummarise aged memory entries into stubs + monthly archive sidecars

Subcommand Details

show [tag]

No arg → print index.md. With tag → filter all files for matching entries. Output table: Date | Title | Tags | File | Tier

search <query>

grep/ripgrep across learnings.md, decisions.md, pitfalls.md, patterns.md, conventions.md. Return file:line + snippet, ranked by relevance.

add <category> <title>

Categories: learning decision pitfall pattern convention Prompt via AskUserQuestion for: what, why it matters, tags (controlled vocab). Append to the matching file using:

### [YYYY-MM-DD] <title>  `[tag1, tag2]`
**What:** ...
**Why it matters:** ...
**Evidence:** ...

That is the whole write. index.md is derived — scripts/memory-index.py rebuilds it at the next session start. Never append index rows by hand; to refresh it now, run python3 scripts/memory-index.py .hyperflow.

edit <entry-id>

Locate by date+title slug. Show current value, prompt for new value, update in place.

prune

Per memory-system.md pruning protocol:

  • Remove [SUPERSEDED] entries older than 7 days
  • Remove entries whose referenced files no longer exist (test -f)
  • Archive entries unreferenced 90+ days to .hyperflow/memory/archive/YYYY-MM.md Print summary of removed/archived counts.
archive

Compress hot entries older than 30 days → .hyperflow/memory/archive/YYYY-MM.md. Leave one-line summary in original file. The stub keeps its date, so the derived index re-tiers it to cold on its own.

clear

Confirm via AskUserQuestion: "This wipes all memory for this project. Are you sure?" If yes → move all content to .hyperflow/memory/archive/cleared-<timestamp>.md, then reset files to empty stubs.

stats

Print: total entries, hot/warm/cold counts, tag frequency table, oldest and newest entry dates.

migrate

Read ~/.claude/hyperflow-memory.md, filter entries matching current project path. Append matching entries to learnings.md. Leave legacy file untouched. Print count of migrated entries.

off

Print: "Memory writes disabled for this session." No files modified.

compact

User-invoked memory compaction. Summarises entries older than 7 days into stub lines and preserves the full text in monthly archive sidecars at .hyperflow/memory/archive/YYYY-MM.md.

Flow:

  1. The compact subcommand handler reads the target memory file (default: learnings.md; pass a path to target another).
  2. The Date/tag parser splits entries into hot (≤7 days, preserved) and eligible (>7 days). Both [domain, type] and legacy backticked `[domain, type]` tag forms are accepted.
  3. The Compaction Writer is dispatched in a single batch with all eligible entries.
  4. The Stub formatter renders each replacement line as ### [YYYY-MM-DD] Short title [domain, type] — summarized, see archive/YYYY-MM.md.
  5. The Dedup Reviewer performs source-side stub-line match and archive-side header match (date + title + tags on both sides) to prevent duplicates.
  6. The Archive-sidecar writer appends accepted entries to archive/YYYY-MM.md, grouped by each entry's calendar month.
  7. The source file is rewritten with stubs replacing the original entries.
  8. The compact subcommand handler refreshes .hyperflow/memory/.checksums (a memory-scoped sidecar — distinct from .hyperflow/.checksums which the scaffold staleness check owns) and exits with a summary.

Output: N entries compacted into archive/YYYY-MM.md · M stubs rejected as duplicates · source N→M lines. Full protocol in compaction.md.

Show full SKILL.md (367 more words)Show less

Flow

  1. Parse invocation to determine subcommand
  2. If subcommand missing → list subcommands table above with one-line descriptions
  3. Execute subcommand
  4. Print structured result with counts/changes summary

Overview

/hyperflow:cache is the operator interface to project-scoped memory under .hyperflow/memory/. It's the only skill that mutates memory files directly (other skills append via the memory-system protocol). Subcommands cover the full lifecycle: show, search, add, edit, prune, archive, clear, stats, migrate. All operations are project-local — entries never leak across projects.

Prerequisites

  • .hyperflow/ initialized (run /hyperflow:scaffold if missing — cache creates .hyperflow/memory/ on first write but expects the parent dir).
  • Write access to .hyperflow/memory/ and .hyperflow/memory/archive/.
  • For migrate only: read access to ~/.claude/hyperflow-memory.md (legacy global memory).

Instructions

See Subcommands and Subcommand Details above for the full operational spec. Summary:

  1. Parse the subcommand from the user's invocation (or list subcommands if none given).
  2. Validate prerequisites for the chosen subcommand (e.g. clear requires AskUserQuestion confirmation; migrate requires legacy file presence).
  3. Execute the subcommand against .hyperflow/memory/.
  4. Print structured result with counts and any file-level changes.

Output

Each subcommand prints a compact summary:

  • show — table of matching entries (Date | Title | Tags | File | Tier).
  • search — file:line matches with snippets, ranked by relevance.
  • add / edit — confirmation line with new entry id and target file.
  • prune / archive / clear — counts of removed/archived/cleared entries plus destination paths.
  • stats — totals + hot/warm/cold breakdown + top-N tags.
  • migrate — count of migrated entries + source legacy file path.
  • off — single-line Memory writes disabled for this session.

Error Handling

FailureBehavior
.hyperflow/memory/ missingAuto-create skeleton (index.md + 5 category files + archive/.gitkeep) on first write; for read-only subcommands, print (no memory yet — invoke /hyperflow:scaffold first).
Subcommand unknownPrint subcommands table; suggest closest match via Levenshtein distance.
add with invalid categoryReject and list valid categories: learning, decision, pitfall, pattern, convention.
edit entry id not foundList 3 closest matches by title slug + date.
clear without confirmation (headless)Refuse and print clear requires interactive confirmation. Do not wipe.
migrate source file missingPrint (nothing to migrate — ~/.claude/hyperflow-memory.md not found) and stop.

Examples

Show all entries
/hyperflow:cache show

Date         Title                              Tags                  File              Tier
2026-05-16   Bash scoping required by validator [validator, marketplace] learnings.md   hot
2026-05-15   No AI attribution in commits       [convention, git]     conventions.md    hot
2026-05-14   Per-task commits in plugin dev     [convention, git]     conventions.md    hot
3 entries (3 hot, 0 warm, 0 cold)
/hyperflow:cache search "validator"

.hyperflow/memory/learnings.md:42 — "Jeremy's validator requires scoped Bash..."
.hyperflow/memory/decisions.md:8 — "...validator score of 73 → 94 after fix"
2 matches
Add a learning
/hyperflow:cache add learning "Markdown frontmatter needs block scalar for colons"

? What: Block scalar (|) preserves : and backticks in YAML values
? Why it matters: prevents fatal YAML parse failures in marketplace validators
? Tags: yaml, validator, frontmatter
Added — .hyperflow/memory/learnings.md (entry 2026-05-16-block-scalar-frontmatter)
Stats
/hyperflow:cache stats

Memory entries: 47
  Hot   (≤7d)   12
  Warm  (8-30d) 23
  Cold  (30d+)  12
Top tags: validator (8), convention (7), git (6), yaml (4)
Oldest: 2026-02-14   Newest: 2026-05-16

Resources

© jeremylongshore, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 4 other files (references) in plugins/ai-agency/hyperflow/skills/cache of jeremylongshore/tons-of-skills-marketplace.

  • SKILL.md
  • references/DOCTRINE.md
  • references/compaction.md
  • references/memory-system.md
  • references/output-style.md

Open the folder on GitHubat commit cfae287

Compare with similar skills

Cache 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 compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Cache this skilljeremylongshore/tons-of-skills-marketplace2.8k—~2.5kAutomated safety check: PassMIT
Prompt Cachingsickn33/agentic-awesome-skills47k2 repos~3.4kAutomated safety check: PassMIT
Prompt Cachingdavila7/claude-code-templates33k5 repos~452Automated safety check: PassMIT
Turborepo Cachingwshobson/agents40k9 repos~2kAutomated safety check: NotesMIT
OmniRoute LLM Cachediegosouzapw/OmniRoute75k—~529Automated safety check: PassMIT
Cachingzebbern/claude-code-guide4.7k—~1.5kAutomated safety check: PassMIT

Similar skills

  • Prompt Caching

    sickn33/agentic-awesome-skills

    Caching strategies for LLM prompts including Anthropic prompt caching, response caching, and CAG (Cache Augmented Generation)

    47k GitHub starsUsed in 2 repos~3.4k tokens
    AI & LLM EngineeringAuto-check passed

  • Prompt Caching

    davila7/claude-code-templates

    Caching strategies for LLM prompts including Anthropic prompt caching, response caching, and CAG (Cache Augmented Generation) Use when: prompt caching, cache prompt, response cache, cag, cache…

    33k GitHub starsUsed in 5 repos~452 tokens
    Backend & APIsAuto-check passed

  • Turborepo Caching

    wshobson/agents

    Configures Turborepo pipelines and local or remote caching for monorepo builds, including Vercel remote cache, a self-hosted cache and cache-miss debugging.

    40k GitHub starsUsed in 9 repos~2k tokens
    DevelopmentAuto-check: notes

  • OmniRoute LLM Cache

    diegosouzapw/OmniRoute

    Documents OmniRoute's cache endpoints for reading cache statistics and clearing entries, statistics or the reasoning cache, with notes on TTL and similarity settings.

    75k GitHub stars~529 tokensUpdated today
    Backend & APIsAuto-check passed

  • Caching

    zebbern/claude-code-guide

    Caching strategies — invalidation, TTL guidelines, cache keys, cache layers, and when not to cache.

    4.7k GitHub stars~1.5k tokensUpdated yesterday
    Backend & APIsAuto-check passed

  • Content-Hash File Cache Pattern

    affaan-m/ECC

    Caches slow file processing results in Python keyed by a SHA-256 hash of the file content, so renames still hit the cache and edits invalidate it automatically.

    277k GitHub starsUsed in 5 repos~1.4k tokens
    DevelopmentAuto-check passed

More from jeremylongshore/tons-of-skills-marketplace

All 3,342 skills in this repo

  • Performing Security Code Review

    jeremylongshore/tons-of-skills-marketplace

    Execute this skill enables AI assistant to conduct a security-focused code review using the security-agent plugin.

    2.8k GitHub starsUsed in 2 repos~1.3k tokens
    Auto-check: notes

  • Adapting Transfer Learning Models

    jeremylongshore/tons-of-skills-marketplace

    Build this skill automates the adaptation of pre-trained machine learning models using transfer learning techniques.

    2.8k GitHub stars~1.1k tokensUpdated today
    Auto-check passed

  • Agent Context Loader

    jeremylongshore/tons-of-skills-marketplace

    Execute proactive auto-loading: automatically detects and loads agents.md files.

    2.8k GitHub stars~1.1k tokensUpdated today
    Auto-check passed

  • Aggregating Performance Metrics

    jeremylongshore/tons-of-skills-marketplace

    Aggregate and centralize performance metrics from applications, systems, databases, caches, and services.

    2.8k GitHub stars~1.2k tokensUpdated today
    Auto-check passed

  • Analyzing Capacity Planning

    jeremylongshore/tons-of-skills-marketplace

    Execute this skill enables AI assistant to analyze capacity requirements and plan for future growth.

    2.8k GitHub stars~947 tokensUpdated today
    Auto-check passed

  • Analyzing Database Indexes

    jeremylongshore/tons-of-skills-marketplace

    Process use when you need to work with database indexing. An agent skill from jeremylongshore/tons-of-skills-marketplace.

    2.8k GitHub stars~2k tokensUpdated today
    Auto-check passed

Questions about Cache

What does Cache do?

A skill your agent uses when the user wants to view, search, add, edit, prune, archive, or clear hyperflow memory entries. Cache is an agent skill from jeremylongshore/tons-of-skills-marketplace. Use when the user wants to view, search, add, edit, prune, archive, or clear hyperflow memory entries.

When should I use Cache?

Cache fits situations like: the user wants to view; clear hyperflow memory entries; with /hyperflow:cache; search memory for X.

How do I install Cache in Claude Code?

Run `npx skills add jeremylongshore/tons-of-skills-marketplace --skill cache -a claude-code`. Or copy the skill folder (plugins/ai-agency/hyperflow/skills/cache in jeremylongshore/tons-of-skills-marketplace) into .claude/skills/cache in your project. Claude Code loads it when a task matches its description.

How do I install Cache in Codex?

Run `npx skills add jeremylongshore/tons-of-skills-marketplace --skill cache -a codex`. Or copy the skill folder (plugins/ai-agency/hyperflow/skills/cache in jeremylongshore/tons-of-skills-marketplace) into .agents/skills/cache in your project. Codex loads it when a task matches its description.

Can I use Cache 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 jeremylongshore/tons-of-skills-marketplace --skill cache -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/cache, .gemini/skills/cache, .github/skills/cache and .opencode/skills/cache in your project.

What does Cache need to run?

Going by SKILL.md and its folder, Cache needs the command-line tools its instructions call (python3). Our summary lists: Python 3. Its frontmatter pre-approves these tools: Read, Write, Edit, Bash(ls:*), Bash(mv:*), Bash(rm:*), Glob, Grep, Agent, AskUserQuestion. Compatibility (from SKILL.md): Designed for Claude Code.

Does Cache 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 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 Cache use?

Cache is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Cache use?

About 2.5k tokens (SKILL.md is roughly 9.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 18k tokens, read only when the agent opens those files.

What are the alternatives to Cache?

Skills that share tags, products or a category with Cache: Prompt Caching (sickn33/agentic-awesome-skills, 47k stars), Prompt Caching (davila7/claude-code-templates, 33k stars), Turborepo Caching (wshobson/agents, 40k stars) and OmniRoute LLM Cache (diegosouzapw/OmniRoute, 75k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Cache?

jeremylongshore (a GitHub user) maintains it in jeremylongshore/tons-of-skills-marketplace, which has 2,827 GitHub stars. The repository holds 3,342 skills in this directory. The repository was last updated on October 10, 2026.

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