Lightweight Q&A about the project — answer from context and return, no routing, no state change.
Scanned 9/2/2026
Install to Claude Code
Skills Directorynpx -y skills add arbiterForge/codeArbiter --skill ca-btw --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Ca Btw?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/arbiterforge-ca-btw-codearbiter)More formats (shields.io, HTML) on the badges page.
---
name: ca-btw
description: Lightweight Q&A about the project — answer from context and return, no routing, no state change.
argument-hint: "<question>"
---
# /ca-btw — quick Q&A
The one exception to the slash-command pipeline. Answer the question and return. No skill is routed
to, no agent dispatched, no file written. Always safe to invoke.
## Flow
1. Read `<project-root>/.codearbiter/` docs as needed to answer accurately — `CONTEXT.md`,
`tech-stack.md`, `coding-standards.md`, `security-controls.md`, `decisions/`, `specs/`, `plans/`.
2. Answer directly and conversationally — not a formatted report.
3. If the answer is not in context, say so plainly. Never guess.
## Hard gate
Read-only. MUST NOT write or modify any file, route to a skill, dispatch an agent, run tests, or
resolve a `[CONFIRM-NN]`. If the question is really a request to build or fix something, say so and
point to `/ca-feature` or `/ca-fix` — do not start implementation from `/ca-btw`.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...
