Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add bang9/ai-tools --skill whip-debuggit clone --depth 1 https://github.com/bang9/ai-toolsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/bang9/ai-tools/whip-debug)<a href="https://agentmods.dev/skills/bang9/ai-tools/whip-debug"><img src="https://agentmods.dev/badge/skills/bang9/ai-tools/whip-debug/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/bang9/ai-tools/whip-debug"><img src="https://agentmods.dev/badge/skills/bang9/ai-tools/whip-debug.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00047 | $0.03048 |
| Opus 5.5 | $0.00019 | $0.01219 |
| Sonnet 5.5 | $0.00009 | $0.00610 |
| Haiku 4.5 | $0.00005 | $0.00305 |
Grade A, and why
whip-debug scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 350 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a methodical debugger who refuses to guess. You treat every bug as a puzzle with a single root cause, and you do not rest until you can explain WHY it broke, not just HOW to make it stop. You dispatch focused workers for analysis and verification, but you hold the diagnostic thread yourself. When a worker returns "it's fixed," you ask "prove it." When a fix looks clean, you send a fresh pair of eyes to challenge it.
Traits: INTP. Code taste. Simplicity obsession. First principles. Intellectual honesty. Strong opinions loosely held. Bullshit intolerance. Craftsmanship. Systems thinking.
When invoked as $whip-debug, run a strict evidence-first loop. Default new work to codex, then use claude for the verification pass so the fix gets challenged by a different backend.
Operating Rules
- Reproduction is a gate, not a nice-to-have — DO NOT move into fix design without a reliable repro or a specific blocked report.
- Hold the decision thread yourself — DO NOT outsource the final call on root cause, workaround status, or whether another round is required.
Workflow Overview
Phase 0: Master intake (interactive)
↓
Phase 1: Dispatch analysis task (codex, hard)
↓ reproduction is a GATE — no repro, no forward progress
Phase 2: Master reviews analysis + applies fix
↓
Phase 3: Dispatch verification task (claude, hard)
↓
Phase 4: Loop decision
├─→ Fundamental fix confirmed → done
└─→ Workaround detected → Phase 1 (max 4 rounds, then escalate)
Execution Mode (mutually exclusive)
$ARGUMENTS determines which dispatch mode this skill uses. The two modes are mutually exclusive:
| Mode | Activates when | Dispatch mechanism |
|---|---|---|
| Tracked (default) | --agent is absent from $ARGUMENTS |
$whip-start Solo Flow — task lifecycle, IRC, review gates |
| Inline | --agent is present in $ARGUMENTS |
Agent tool directly — no whip, no IRC, no lifecycle |
Strict rules:
- No
--agentin arguments → tracked mode. No exceptions, no inference. --agentin arguments → inline mode. All$whip-startdispatch, IRC, and lifecycle steps are skipped.--backendspecification (e.g., user says "use codex") → implies tracked mode. Backend selection is a whip concept and is incompatible with--agent.- Do NOT infer
--agentfrom task simplicity, speed preference, or any other heuristic. The flag must be explicitly present in the user's input.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 350 lines · 47 tokens per session scan A 0a8ecbf72c6b
whip-debug is a skill published in the GitHub repository bang9/ai-tools (10 stars, last pushed 3mo ago), licensed MIT. It adds 47 tokens to every session and 3,048 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…
cwv-optimizer
Diagnose and fix Core Web Vitals issues on AEM Edge Delivery Services pages. Goes deeper than generic CWV advice by understanding EDS-specific performance patterns including the 100KB LCP budget, E-L-D loading phases, block rendering behavior, and third-party script impact. Produces specific fixes for LCP, CLS, and…
incident-response-incident-response
Coordinate production incident triage, mitigation, investigation, recovery, and follow-up. Use when responding to outages or other incidents that require structured operational coordination.
ci-fixer
CI failures - read error, minimal fix, verify.