agent-tools skills

944 tagged agent-tools, measured the same way as everything else here.

Browse within: agent 356ai-skills 272financial-data 218ftshare 218market-data 218quantitative-finance 218agentskills 201cli 198LLM 127knowledge-graph 91ai-video 73claude-code-skills 73api 69graphql 66

clawdrive

49

Hyper3Labs/clawdrive

Skill Claude CodeCodex needs its repo

ClawDrive — agent-native local file platform with multimodal semantic search. Use for adding files, searching, sharing pots, inspecting/downloading shares, transcription (see transcription.md), and tunnels (see tunnels.md).

not rated 5 5mo ago A 47 tokens original MIT

code-review

50

lzehrung/codegraph

Skill Claude CodeCodex

Review changes against this repository's agent directives.

not rated 5 +1 today A 12 tokens original MIT

clrun

51

cybertheory/clrun

Skill Claude CodeCodex

Run and control interactive CLI sessions for AI agents. Handles TUI prompts (select lists, checkboxes, confirms), persistent shell state, and long-running processes. Use when you need to execute terminal commands, respond to interactive prompts, navigate scaffolding wizards like create-vue or create-vite, or manage…

not rated 4 6mo ago A 68 tokens original MIT

wrench

52

hraness/wrench

Skill Codex

Use Wrench to give a bring-your-own coding agent bounded, local-first web capabilities: capture or scrape public and signed-in web pages into Markdown; archive authorized audio, video, and transcripts with verification; query encrypted cached email, contacts, inbox, and messaging views; resolve a live conversation…

not rated 4 changed yesterday A 239 tokens original MIT

code-reviewer

54

osouthgate/agent-plus

Skill Claude Code

Reviews code for quality, bugs, security, and best practices. Use when changing code or before merge. Checks security, authentication, and validation; ensures new code follows project rules and conventions.

not rated 4 2mo ago A 42 tokens original MIT

xiexie-qiuligao/agent-evidence-mcp

Skill Claude CodeCodex

Capture screenshots, short recordings, and milestone evidence during long-running agent tasks. Use when an agent is asked to perform multi-step browser, desktop, QA, troubleshooting, deployment, or admin workflows where the user wants checkpoint artifacts, progress evidence, error snapshots, or a final timeline of…

not rated 4 4mo ago A 65 tokens original MIT

hello-world

56

VictorTomaili/skill-cli

Skill Claude CodeCodex

Minimal example skill — a copy-paste template for your own skills.

not rated 4 today A 17 tokens original MIT

avvarre:reviewer

57

PralhadYadawad/avvarre

Skill Codex

Perform a severity-first Avvarre code review and verify every requested fix.

not rated 4 11d ago A 22 tokens original MIT

anywrite

58

Antheurus/anywrite

Skill Claude CodeCodex

Use this skill whenever the user mentions "anywrite" by name, or mentions Anytype, asks to create/update/search/organize notes, tasks, or PKM objects in Anytype, or wants to upload files, manage properties/tags/types, or chat inside an Anytype space. Covers all 52 endpoints of the Anytype local API (spaces, objects…

not rated 3 1mo ago A 115 tokens original MIT

autoworks-ai/autovault

Skill Codex

Install the AutoHub docs-drift Codex automation from signed AutoVault skill resources and audit its render fidelity with autovault doctor.

not rated 3 3d ago A 36 tokens original MIT

ui-context

60

Fish121380/auto-mouse

Skill Claude CodeCodex

A tool for reading the desktop window, interface element or screen area that a user selects with Little Mouse Pointer.

not rated 3 18d ago A 26 tokens original MIT

frugal

61

brainsparker/frugal

Skill Claude Code

When the Frugal MCP server is connected (tools prefixed frugal), prefer its routed equivalents over native built-in tools. Frugal routes each call to a provider per the operator's routing policy (cheapest-first by default, with automatic failover) and surfaces the provider and cost in the response, so the routing…

not rated 2 3d ago C 79 tokens

zerodust

62

andresdefi/zerodust

Skill Claude CodeCodex

Sweep 100% of native gas tokens (ETH, BNB, MATIC, etc.) from EVM chains via ZeroDust, leaving exactly zero balance. Use when: user wants to exit a chain completely, consolidate dust balances, clean up wallets, or bridge remaining native tokens cross-chain. Supports 25 mainnet chains including Ethereum, Base, Arbitrum…

not rated 2 1mo ago A 93 tokens original MIT

searchts

63

capad-xyz/searchts

Skill Claude CodeCodex

MUST USE for web research, lookup, or any shared URL/link (Twitter/X, Reddit, YouTube, GitHub, LinkedIn, or any page). Read with searchts read — unlocker for 403/429/bot-walls and JS. Search with searchts search. Transcribe with searchts transcribe. Grab assets/palette with searchts grab or searchts get. Prefer…

not rated 2 yesterday A 110 tokens original MIT

belief-assessor

64

hqzzdsda/belief-state-runtime

Skill Claude CodeCodex

LLM-driven epistemic reasoning engine. Evaluates claims against evidence, outputs calibrated confidence and structured belief state (VERIFIED/CONTESTED/UNCERTAIN). v2 adds 4-way constraint system, parameterized configuration, and formula-based confidence intervals. Use when the agent needs to assess whether…

not rated 2 2mo ago A 77 tokens

planmaxx

65

AlhasanIQ/planmaxx

Skill Claude CodeCodex

Use when an agent-written plan, design, or spec needs user review before implementation, or when the user asks to review, approve, or iterate on a plan.

not rated 2 21d ago A 37 tokens GPL-3.0

ixf-docs-writer

66

serialq7ic4/ixf-toolbox

Skill Claude CodeCodex

Use when publishing local Markdown as a new i讯飞 docx document or updating an approved existing docx document.

not rated 2 10d ago A 30 tokens

ctx2img

67

Koukyosyumei/h5i-ctx2img

Skill Claude CodeCodex

Cut context tokens by rendering it as images. ctx2img paint turns any text-shaped input (a repo, a directory, a file, markdown, stdin) into dense images that carry the full text at 60-75% fewer tokens, with stable handles and a verbatim factsheet; ctx2img read recovers guaranteed-exact text. Use before ingesting any…

not rated 2 1mo ago A 103 tokens original Apache-2.0

freeact

68

xuviga/freeact

Skill Claude CodeCodex

Skill "freeact" from xuviga/freeact, covering freeact — the browser tamer, the sacred cycle, quick start, configuration and part i: browser automation patterns.

not rated 2 1mo ago A 0 tokens

data-analysis

69

furkangonel/cowrangler

Skill Claude CodeCodex

Structured data analysis workflow from raw data to shareable insights.

not rated 2 yesterday A 15 tokens original MIT

planpage

70

YosefHayim/planpage

Skill Claude CodeCodex

Render a skill's plan, review gate, quiz, poll, or report as a beautiful, self-contained INTERACTIVE HTML page via the open-source planpage package (Preact → static HTML + local post-back). The user edits and annotates the plan in the browser, attaches screenshots, stages everything in a fixed sidebar, then hits Send…

not rated 2 1mo ago A 123 tokens original MIT

notify-user

71

5dive-ai/5dive-plugins

Skill Claude CodeCodex

Telegram comms playbook for Antigravity — when and how to reply, ack long work, present choices, and delegate to other agents. Use whenever you're paired to a Telegram chat via the telegram-agy MCP server and need to talk back to the user. The chat is your only signal channel — terminal output never reaches the user.

not rated 2 3d ago A 71 tokens original Apache-2.0

commune-email

72

black22345/commune-cookbook

Skill Claude CodeCodex

I give you a real email inbox and let you send, receive, search, and manage email threads programmatically via the Commune API. Every inbox I create is RFC 5322 compliant — replies thread correctly in any email client, including Gmail, Apple Mail, and Outlook.

not rated 2 2d ago A 0 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: