Ada — Pair Programmer
Persona
Ada is the second set of eyes that doesn't flinch — the programmer who reads your diff like a reviewer with a stake in the outcome.
About
Stop shipping code that "looks fine" to an AI that never actually read it
Ada is the pair programmer persona for OpenClaw — reads diffs carefully, flags blocking issues by name, and pushes back on shortcuts exactly once before deferring to your call. What makes Ada distinct: she doesn't say 'Looks good!' without telling you what she checked. She won't hallucinate an API or invent a function signature to sound confident. If she's not sure, she says 'check the docs on this' and stops.
What's inside
- SOUL.md — a coder's voice, five values ranked, 50-word-per-comment discipline
- HEARTBEAT.md — morning check-in: open PRs, failing tests, one technical risk from yesterday
- AGENTS.md — two sub-agents: code-reviewer (blocking vs. non-blocking, 5 items max) and pr-scoper (split big branches into reviewable PRs)
- TOOLS.md — confirm-first defaults: no silent file writes, no silent commits, no silent installs
- /code-reviewer skill — structured review every time, reads the whole thing, groups by severity
- /pr-scoper skill — splits large branches into 2-4 PRs with merge order and separation rules
- Setup Guide PDF — drop-in install instructions
Who this is for
- Solo developers who want a second set of eyes on every diff
- Developers on small teams where PR review is slow or irregular
- Anyone who catches their own bugs only after shipping them
- OpenClaw users who want a personality with actual opinions
How it ships
Drop the files into ~/.openclaw/, restart, done. Morning check-in fires with your open PR context already loaded. Paste code, get a numbered review with blocking issues at the top.
Built and refined through five convergence runs.
Questions: hackedcubeapps@gmail.com (response within 48 hours).
Core Capabilities
- Reviews diffs with numbered, severity-grouped issues capped at five items
- Splits large branches into two to four reviewable PRs with explicit merge order
- Keeps review comments under 50 words each — if it takes more, she's not sure what's wrong
- Refuses silent file writes, silent commits, and silent installs
- Flags API uncertainty rather than fabricating function signatures
- Pushes back on shortcuts once with reasoning, then helps you ship it your way
- Surfaces yesterday's risk and today's failing tests in a morning check-in
Customer ratings
0 reviews
No ratings yet
- 5 star0
- 4 star0
- 3 star0
- 2 star0
- 1 star0
No reviews yet. Be the first buyer to share feedback.
Version History
This persona is actively maintained.
April 18, 2026
Initial release
One-time purchase
$29
By continuing, you agree to the Buyer Terms of Service.
Details
- Type
- Persona
- Category
- Engineering
- Price
- $29
- Version
- 1
- License
- One-time purchase
Works With
Requires OpenClaw runtime features.
Recommended Skills
Skills that complement this persona.
Qwen3 6-35B-A3B released
Engineering
Remember the people who matter — before it's too late
$14
Persistant memory upgrade
Engineering
Your agent remembers who you are, what you're building, and how you work — across every session, forever.
$5
Release Notes from PRs
Engineering
Generate user-facing release notes from merged PRs — auto-categorized by type, rewritten from commit-speak into changelog English, and formatted for Keep-a-Changelog.
$29