Kiro Prompt Pack

A collection of 12 reusable prompts for the Kiro CLI.

Installation

Option 1: Clone for a single project (local prompts)

# In your project root
git clone https://github.com/llamojha/kiro-prompts.git .kiro-prompts-temp
cp -r .kiro-prompts-temp/.kiro/prompts .kiro/prompts
rm -rf .kiro-prompts-temp

Option 2: Install globally (available in all projects)

git clone https://github.com/llamojha/kiro-prompts.git
cp -r kiro-prompts/.kiro/prompts ~/.kiro/prompts
rm -rf kiro-prompts

Option 3: Cherry-pick individual prompts

# Copy just the prompts you want
curl -o ~/.kiro/prompts/code-review.md https://raw.githubusercontent.com/llamojha/kiro-prompts/main/.kiro/prompts/code-review.md

Usage

In Kiro CLI chat, invoke any prompt with @prompt-name:

@code-review
@security-checklist
@debug-helper

Prompts

commit-message

Generate a commit message for staged changes.

• Input: Auto-scans git staged changes
• Format: Conventional Commits (feat/fix/docs/style/refactor/test/chore)
• Output: One-line summary (max 72 chars), optional body for complex changes

---

code-review

Review staged changes for correctness, security, reliability, readability, and test coverage.

• Input: Auto-scans git staged changes
• Output: Issues with severity, location, problem description, and fix suggestions

---

mock-testing

Generate a test plan with mocks/stubs strategy and sample tests.

• Input: Paste code or reference with @filename
• Framework: Auto-detects from project config (package.json, pyproject.toml). Asks if not found.
• Output: Test plan, mocking strategy, and working test implementations

---

daily-next-step

Summarize current status and output next actions for today.

• Input: Auto-scans .kiro/specs/ and TODO.md
• Output: Status summary (done/blocked/in-progress) and 1-3 concrete next actions

---

security-checklist

Security-focused review of staged changes.

• Input: Auto-scans git staged changes
• Reviews: Secrets, auth/authz, input validation, least privilege, sensitive data logging, dependency risks
• Output: Findings with risk level and mitigation recommendations

---

debug-helper

Debug helper for errors, stack traces, and logs.

• Input: Paste error, or auto-scans terminal output and log files (*.log, logs/, .kiro/logs/, tmp/)
• Output: Likely root causes, clarifying questions, debugging actions, safe fix with rollback notes

---

prompt-builder

Turn a rough goal into a polished, reusable prompt.

• Input: Describe your rough goal or idea
• Output: Clarifying questions, then a polished prompt with clear objective, constraints, and output format

---

refactor-code

Refactor staged changes for clarity, structure, and performance.

• Input: Auto-scans git staged changes
• Output: List of changes with rationale, before/after diff, verification steps

---

generate-documentation

Generate documentation for code.

• Input: Reference file/folder with @filename, paste code, or say "staged changes"
• Doc style: Auto-detects (JSDoc, docstrings, etc.) or specify preferred format
• Output: Overview, API docs, examples, edge cases and gotchas

---

codebase-search

Find or change code in the codebase.

• Input: Describe what you want to find or change
• Output: Search targets (symbols, patterns, regex), likely locations, step-by-step execution plan

---

seo-meta-sitemap-robots

Generate or improve SEO and crawler assets.

• Input: Auto-scans existing SEO files. Asks for URL if not found.
• Output: Social tags (Open Graph/Twitter), sitemap.xml, robots.txt, llms.txt with exact file edits

---

refresh-foundational-steering-and-docs

Refresh foundational steering documents and docs.

• Input: Auto-scans .kiro/steering/, docs/, and README.md
• Output: Audit of outdated guidance, revised structure, updated conventions, PR checklist

Quick Reference

Prompt	Input	Description
commit-message	Auto: staged changes	Generate conventional commit message
code-review	Auto: staged changes	Review for correctness, security, reliability, readability
mock-testing	Code input	Generate test plan with mocks/stubs
daily-next-step	Auto: specs + TODO.md	Status summary and next actions
security-checklist	Auto: staged changes	Security review
debug-helper	Paste or auto: logs	Root cause analysis and fix suggestions
prompt-builder	Describe goal	Turn rough goals into polished prompts
refactor-code	Auto: staged changes	Refactor with before/after diff
generate-documentation	File, paste, or staged	Generate docs
codebase-search	Describe target	Search targets and execution plan
seo-meta-sitemap-robots	Auto or URL input	SEO and crawler assets
refresh-foundational-steering-and-docs	Auto: steering + docs	Refresh steering docs

Agents

rubberduck

A thinking partner for debugging and brainstorming using the Rubberduck method.

• Usage: Switch to agent with /agent rubberduck
• Capabilities: Asks clarifying questions, helps break down problems, guides you to solutions through conversation
• Tools: Read-only file access (can explore your codebase but won't make changes)

Contributing

Contributions welcome! Feel free to:

• Submit new prompts via PR
• Improve existing prompts
• Report issues

License

MIT License - see LICENSE for details.