turnstile-systematic-debugger

A step-by-step method for finding the confirmed cause of a software bug. It requires reproducing the problem, collecting evidence, testing a specific explanation, and then routing the fix.

In plain words
What is it for?
Use it to investigate bug reports, inspect related notes and code changes, verify a root cause, and pass a confirmed minimal fix to the coding agent.
Why use it?
It helps prevent guesses and patches that only hide symptoms when the real cause is unknown.

Skill for Claude CodeCodex

Part of the turnstile plugin — 24 skills, 23 commands, 5 agents, 3 hooks, 1 MCP server shipped together

Install

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.

agentmods
npx agentmods add skills/febradc-github/turnstile/turnstile-systematic-debugger
Any agent
npx skills add febradc-github/turnstile --skill turnstile-systematic-debugger
Clone the repo
git clone --depth 1 https://github.com/febradc-github/turnstile

Made for: Claude Code, Codex.

Or install turnstile, the plugin that ships this one along with the rest of its 24 skills, 23 commands, 5 agents, 3 hooks, 1 MCP server.

Per session 52 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 899 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00052 $0.00899
Opus 5 $0.00026 $0.00449
Sonnet 5 $0.00010 $0.00180
Haiku 4.5 $0.00005 $0.00090

Measured 2d ago against content hash ad66f6a34c94, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

turnstile-systematic-debugger 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 2d 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.

skills/turnstile-systematic-debugger/SKILL.md · 31 lines

How it starts

The opening of the file, as written. The whole thing — 31 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Systematic Debugger

Process

  1. Search the vault for notes related to the bug's topic or symptoms ($ARGUMENTS). Surface anything relevant, including a prior note documenting this exact cause.
  2. Reproduce the bug. If it can't be reproduced, gather evidence (error messages, logs, exact steps tried) -- and if none is available, tell the user what's needed before continuing. Never guess at a fix for a bug that hasn't been observed.
  3. Gather evidence: the relevant code paths, recent changes (git log/git diff), error output in full.
  4. Form one specific, falsifiable hypothesis about the root cause.
  5. Test it with the cheapest check available (targeted log, minimal reproduction, reading the exact line) before writing any fix. Wrong hypothesis: discard it and form a new one from the evidence -- never patch around a disproven hypothesis "just in case".
  6. Once the root cause is confirmed, route the fix:
    • Related to the in_progress item (the bug lives in that ticket's scope -- same feature or files): the fix is part of that ticket. Dispatch turnstile-coder with the confirmed root cause, reproduction steps, and intended minimal fix; the fix rides that ticket's diff into /turnstile:review. Append to the item's notes: debug: <one-line root cause and fix>.
    • Unrelated (or nothing in_progress), turnstile board present: do not fix it under someone else's diff. Invoke turnstile-quick with the bug -- title, confirmed root cause, reproduction, intended fix, tagged bug, assignee claude. If an unrelated item is in_progress, the bug task queues behind it; if nothing is, invoke turnstile-work with the new id so the fix happens now. Either way it ships through work -> review under its own ticket.
    • No turnstile board: ad hoc mode -- dispatch turnstile-coder directly with the root cause and fix approach.
  7. Where the coder was dispatched here (related or ad hoc branch): verify the fix resolves the original reproduction case and check for regressions. If it doesn't resolve it, the hypothesis was wrong -- revert and re-investigate rather than layering another change on top. (A queued bug task is verified later by its own work/review cycle.)
  8. Capture (per the capture config), and in either mode only after the root cause is confirmed -- never dispatch mid-investigation, and never pass unconfirmed hypotheses. In gates mode (default), dispatch brain-curator with bounded input: the confirmed root cause (plus the touched files -- path, purpose, what changed -- when the fix changed code here). In opportunistic mode, dispatch when the fix changed code or the root cause was non-obvious; one dispatch covers both.
  9. Tell the user the root cause and what happened to the fix: folded into the in_progress ticket, queued/fixed as ticket <id>, or applied ad hoc.

Read the full file on GitHub · 31 lines

Changes

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.

  1. 2d ago First seen · 31 lines · 52 tokens per session scan A ad66f6a34c94

Subscribe to this mod's changes

turnstile-systematic-debugger is a skill published in the GitHub repository febradc-github/turnstile (1 stars, last pushed 1mo ago), licensed MIT. It adds 52 tokens to every session and 899 once invoked, about $0.0003 per session on Opus 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.

Related

Other skills, from other repositories

model-compatibility

Model family compatibility matrix covering loaders, resolutions, samplers, CFG, VAE, ControlNet, and LoRA compatibility for SD 1.5, SDXL, Flux, SD3, and video models.

artokun/comfyui-mcp · 47 tokens

comfyui-node-registry

Authoring & publishing ComfyUI custom nodes to the Comfy Registry, covering node structure, pyproject.toml spec, comfy-cli publishing, and CI.

artokun/comfyui-mcp · 38 tokens

flux-txt2img

Build Flux txt2img workflows with Flux.1 Dev (SRPO), Flux 2 Klein 9B, Turbo LoRAs, FluxGuidance, and DualCLIPLoader patterns.

artokun/comfyui-mcp · 44 tokens

civitai

Discover Civitai models with the BUILT-IN downloadmodel action:"searchcivitai" and install/generate them locally. Find a checkpoint/LoRA/embedding on Civitai, download it into ComfyUI, and use its trigger words. Optionally pair the official Civitai MCP for community features (images browsing, posting, collections).

artokun/comfyui-mcp · 76 tokens

chief-strategy-officer

Owns where the business plays and how it wins over a multi-year horizon — portfolio choices, corporate development, strategic partnerships, and planning under uncertainty. Use this for a decision about which markets or businesses to be in, whether to build, buy, or partner, how to allocate capital across business…

cbrock84/headcount · 93 tokens

performance-analysis

Measurement approaches, profiling patterns, bottleneck identification, and optimization guidance. Use when diagnosing performance issues, establishing baselines, identifying bottlenecks, or planning for scale. Always measure before optimizing.

rsmdt/the-startup · 42 tokens