skill

behavioral-modes — Claude Code skill: adaptive AI modes

Q: How do I install behavioral-modes in Claude Code?

Run `npx claude-code-templates@latest --skill ai-research/behavioral-modes`. The skill lands in `~/.claude/skills/behavioral-modes/SKILL.md`.

Q: Can I pick a mode myself?

Yes, ask for one in your message, for example `/debug why login fails`. Without an explicit command the agent infers the mode from trigger words.

Gives the agent a set of operating modes, from brainstorming to shipping. Useful when replies should match the kind of task at hand.

Install into Claude Code · claude-code-templates All methods and platforms ↓

npx claude-code-templates@latest --skill ai-research/behavioral-modes

Open on GitHub

What behavioral-modes does

The skill lays out seven modes: BRAINSTORM, IMPLEMENT, DEBUG, REVIEW, TEACH, SHIP and EXPLORE. Each one spells out how the agent tackles a task, what it asks first and how it formats the answer. IMPLEMENT, for instance, favours finished code with no walkthrough, while DEBUG moves from hypothesis to test to root cause.

Mode selection runs on trigger words. Ask for ideas or options and you get BRAINSTORM; mention a bug or an error and DEBUG kicks in; say deploy and SHIP takes over. You can also switch by hand with /brainstorm, /implement, /debug or /review at the start of a message. The skill only carries Read, Glob and Grep, so it reads your code but never edits it.

Heavy tasks can follow the PLAN-EXECUTE-CRITIC loop: a planner breaks the work into steps in task.md, an executor writes the code, a critic checks it. Install with npx claude-code-templates@latest --skill ai-research/behavioral-modes.

What it helps with

get code instead of a lecture from the agent
trace a bug back to its cause rather than patch symptoms
collect several solution options before committing to one
run a review with findings sorted by severity
work through a pre-release checklist before deploying

Example prompts for the agent

Switch to brainstorm mode and give me architecture options for an orders service
Debug mode: why does auth fail after the token refresh
Review this pull request and sort the findings by severity

How to install behavioral-modes

1claude-code-templates

npx claude-code-templates@latest --skill ai-research/behavioral-modes

2Skills CLI (Vercel)

npx skills add davila7/claude-code-templates --skill behavioral-modes -a claude-code

3Manually: put the files in place

~/.claude/skills/behavioral-modes/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/behavioral-modes/

Manually: put the files in place

.agents/skills/behavioral-modes/SKILL.md

For all projects use ~/.agents/skills/

Manually: put the files in place

.cursor/skills/behavioral-modes/SKILL.md

Cursor also reads .agents/skills/ and .claude/skills/

1GitHub CLI

gh skill install davila7/claude-code-templates behavioral-modes

2Manually: put the files in place

.github/skills/behavioral-modes/SKILL.md

Platform CLI

gemini skills install https://github.com/davila7/claude-code-templates --path cli-tool/components/skills/ai-research/behavioral-modes

Manually: put the files in place

.devin/skills/behavioral-modes/SKILL.md

The legacy .windsurf/skills/ path still works

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about behavioral-modes

How do I install behavioral-modes in Claude Code?

Run npx claude-code-templates@latest --skill ai-research/behavioral-modes. The skill lands in ~/.claude/skills/behavioral-modes/SKILL.md.

What does behavioral-modes do?

It maps task types to agent behaviour: brainstorm, implement, debug, review, teach, ship and explore. Each mode changes the reply style and the order of steps.

Can I pick a mode myself?

Yes, ask for one in your message, for example /debug why login fails. Without an explicit command the agent infers the mode from trigger words.

thinking-model-router tjboudreaux A router skill that maps domain and problem type to one thinking skill, or returns NONE. Useful when you would otherwise guess or stack frameworks. CX Codex +1 605installs goal-plan ruvnet A skill that builds step-by-step plans for complex tasks, tracks what conditions must be met before each step, estimates costs, and switches to an alternative path if something fails. CX Codex +1 770installs nerd-smart Danangjoyoo A skill that pins down the real goal of a request and picks a single route: discuss, explore, diagnose, plan or execute. CX Codex +1 936installs creative-writing-modes haowjy A skill that adds five prose modes to your writing pass: fresh draft, revision, bridge, alternate take and line polish. Useful once structure is settled and the page needs words. CX Codex +1 534installs pixijs-blend-modes pixijs A skill that tells your agent how to composite PixiJS v8 display objects with blendMode: standard modes, advanced ones, and batch-friendly ordering. CX Codex +1 5,700installs cc-best-practices dboeckli A skill that tells your agent how to work with Claude: context management, plan mode, verification, and the failure patterns to avoid. CX Codex +1 3,500installs

From the same set495

The whole set →

git-bisect-helper Runs a git bisect session and marks commits good or bad from your test's exit code. Useful when you know something broke but not where. 163installs pac-update-status Moves a Product as Code ticket to a new status, reassigns it, or logs a progress note. Handy when your tickets live in .pac/ and you want to manage them without leaving the terminal. 160installs astro This skill guides Astro projects with Markdown or MDX content and selectively hydrated interactive components. Use it for blogs, documentation, and portfolio sites. 157installs cloud-architect A Claude skill for designing infrastructure across AWS, Azure and GCP, with Terraform, CDK, cost work and security controls. 157installs peer-review Review scientific manuscripts and grant proposals with checks for methods, statistics, reproducibility, ethics, and reporting. Use it to structure a critical review across research fields. 154installs data-engineer This skill helps plan data pipelines, warehouses, and streaming systems. Use it when a project needs ingestion, orchestration, or data quality checks. 152installs pubmed-database A Claude skill for PubMed access over the E-utilities REST API: MeSH queries, batch fetches, citation matching by PMID or DOI. 149installs prompt-engineering-dspy Learn to structure language-model tasks with DSPy, assemble RAG workflows, and tune modules from examples. Useful when a project needs more than a hand-written prompt. 147installs