Skills by risk · page 11

Highest-trust agent skills

7,952 skills rated SAFE with grade A or B, best score first. All are grade B today: grade A needs a sandboxed run we do not do for skills yet. Ties are broken by stars, so a repository with many skills can fill a page.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “Highest-trust agent skills”. Retrieved 29 Oct 2026. https://modelsatlas.com/skills/trusted/page/11
7,952 SAFE skills · page 11 of 332 · ranked by trust score, then stars

Cost Aware Llm Pipeline

affaan-m/ECC

✓SAFE B

LLM APIの使用量のコスト最適化パターン — タスクの複雑さによるモデルルーティング、予算追跡、リトライロジック、プロンプトキャッシング。

★ 262k89/100

Cost Tracking

affaan-m/ECC

✓SAFE B

ローカルのコスト追跡データベースからClaude Codeのトークン使用量、支出、予算を追跡・レポートします。コスト、支出、使用量、トークン、予算、またはプロジェクト、ツール、セッション、日付によるコスト内訳について質問する場合に使用します。

★ 262k89/100

Council

affaan-m/ECC

✓SAFE B

曖昧な決定、トレードオフ、ゴー/ノーゴーの判断のために4つの声のカウンシルを召集します。複数の有効なパスが存在し、選択前に構造化された異議が必要な場合に使用します。

★ 262k89/100

Cpp Coding Standards

affaan-m/ECC

✓SAFE B

C++コアガイドラインに基づくC++コーディング標準(isocpp.github.io)。現代的で安全で慣用的なプラクティスを強制するためにC++コードを書き、レビュー、またはリファクタリングする場合に使用します。

★ 262k89/100

Cpp Testing

affaan-m/ECC

✓SAFE B

C++ テストの作成/更新/修正、GoogleTest/CTest の設定、失敗またはフレーキーなテストの診断、カバレッジ/サニタイザーの追加時にのみ使用します。

★ 262k89/100

Verify

facebook/react

✓SAFE B

Use when you want to validate changes before committing, or when you need to check all React contribution requirements.

★ 252k89/100

Test

facebook/react

✓SAFE B

Use when you need to run tests for React core. Supports source, www, stable, and experimental channels.

★ 252k89/100

Flow

facebook/react

✓SAFE B

Use when you need to run Flow type checking, or when seeing Flow type errors in React code.

★ 252k89/100

Flags

facebook/react

✓SAFE B

Use when you need to check feature flag states, compare channels, or debug why a feature behaves differently across release channels.

★ 252k89/100

Fix

facebook/react

✓SAFE B

Use when you have lint errors, formatting issues, or before committing code to ensure it passes CI.

★ 252k89/100

Feature Flags

facebook/react

✓SAFE B

Use when feature flag tests fail, flags need updating, understanding @gate pragmas, debugging channel-specific test failures, or adding new flags to React.

★ 252k89/100

Extract Errors

facebook/react

✓SAFE B

Use when adding new error messages to React, or seeing "unknown error code" warnings.

★ 251k89/100

Blackbox

NousResearch/hermes-agent

✓SAFE B

Delegate coding tasks to the Blackbox AI multi-model CLI.

★ 248k89/100

Blender Mcp

NousResearch/hermes-agent

✓SAFE B

The agent that grows with you

★ 248k89/100

Evm

NousResearch/hermes-agent

✓SAFE B

Read-only EVM client: wallets, tokens, gas across 8 chains.

★ 248k89/100

Honcho

NousResearch/hermes-agent

✓SAFE B

Configure and troubleshoot Honcho memory for Hermes.

★ 248k89/100

Hyperframes

NousResearch/hermes-agent

✓SAFE B

Render MP4/WebM videos from HTML compositions.

★ 248k89/100

Kanban Video Orchestrator

NousResearch/hermes-agent

✓SAFE B

Plan and run multi-agent video production pipelines.

★ 248k89/100

One Three One Rule

NousResearch/hermes-agent

✓SAFE B

1-3-1 decision briefs: problem, three options, one pick.

★ 248k89/100

Openhands

NousResearch/hermes-agent

✓SAFE B

Delegate coding to OpenHands CLI (model-agnostic, LiteLLM).

★ 248k89/100

Solana

NousResearch/hermes-agent

✓SAFE B

Query Solana wallets, tokens, txs, and NFTs in USD.

★ 248k89/100

Karpathy Guidelines

multica-ai/andrej-karpathy-skills

✓SAFE B

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

★ 221k89/100

Karpathy Guidelines

forrestchang/andrej-karpathy-skills

✓SAFE B

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

★ 221k89/100

Update Screenshots

microsoft/vscode

✓SAFE B

Update the committed blocks-ci screenshot hashes after the "Screenshots & Tests" check fails, or investigate a screenshot diff reported on a PR. Use when asked to update, accept, or refresh component screenshot…

★ 194k89/100