Thinking Tools
This "Thinking Tools" skill provides structured prompts and frameworks to enhance your reasoning & problem-solving abilities for clearer thought processes.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add thinking-tools npx -- -y @trustedskills/thinking-tools
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"thinking-tools": {
"command": "npx",
"args": [
"-y",
"@trustedskills/thinking-tools"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
This skill, "Thinking Tools," provides structured thinking patterns to guide an AI agent's reasoning and problem-solving process. It offers two specific frameworks – “Think About Collected Information” and “Think About Task Adherence” – designed to improve task quality and prevent common errors like incomplete research or scope creep. These patterns involve self-assessment questions and a decision-making step to ensure the agent stays on track.
When to use it
This skill is beneficial in these situations:
- After completing a sequence of searches, such as reading multiple files or exploring directory structures.
- Before making any code changes (insertions, replacements, deletions).
- When gathering requirements for a task.
- To ensure the agent's actions remain aligned with the original request and avoid unnecessary modifications.
Key capabilities
- Information Completeness Check: Evaluates if sufficient, relevant information has been gathered, identifies gaps in understanding, and assesses confidence levels.
- Task Adherence Check: Ensures planned code changes directly address the original task and avoids scope creep or unrelated improvements.
- Structured Self-Assessment Questions: Provides specific questions to guide reflection on collected information and alignment with the initial goal.
- Decision Framework: Offers clear decision options (Proceed, Gather More Information, Ask User) based on self-assessment results.
Example prompts
Here are some example prompts for an agent using this skill:
- "Before you make any code changes, run a 'Think About Task Adherence' check."
- "After reading these files, please perform an 'Information Completeness Check'."
- “What is the original task and how does your current action align with it?”
Tips & gotchas
- The skill is most effective when used proactively at key points during a task.
- Pay close attention to the "Scope Creep Warning Signs" within the “Task Adherence Check” to avoid drifting from the initial request.
- The self-assessment questions are designed to be thorough; taking the time to answer them carefully will improve the agent's performance.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.