Skeptical Triage
3-round self-challenge plus arbiter that catches false-positive review findings before they block a gate.
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 15, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · fb414aa
Fed it a security finding whose only cited defense was a docstring reference to a middleware file that doesn't exist anywhere in the codebase -- the skill's grep-the-defense-or-it-doesn't-exist rule caught it and kept the finding VALID, exactly where an unstructured review would likely have waved it through on the docstring's word.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 5/5
What Skeptical Triage does
Skeptical Triage is a structured 4-step process (3 skeptical review rounds + an impartial arbiter) for deciding whether a code-review, security, or QA finding is a real blocker or a false positive. It triggers when a P0/P1 finding could block a ship/code/QA/arch gate or when two reviewers disagree, and explicitly excludes hard findings (secrets, confirmed CVEs) and P2/advisory notes. In testing it caught an unverifiable 'middleware handles this' defense claim that an unstructured review would likely have accepted at face value.
How to install Skeptical Triage
git clone https://github.com/avelikiy/great_cto
cd great_cto
mkdir -p ~/.claude/skills
cp -r skills/skeptical-triage ~/.claude/skills/skeptical-triage
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Skeptical Triage
-
/skeptical-triage3-round self-challenge plus arbiter that catches false-positive review findings before they block a gate.
It also activates on plain-language prompts like these:
-
is this security finding a real blocker or just a false positive -
two reviewers disagree on this vulnerability, help me arbitrate it -
double check this P1 finding before it blocks our ship gate
Frequently asked questions
- Is the Skeptical Triage skill free?
- Yes. The skill itself is free from avelikiy/great_cto. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Skeptical Triage work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 15, 2026. Verdict: Tested · Works. Fed it a security finding whose only cited defense was a docstring reference to a middleware file that doesn't exist anywhere in the codebase -- the skill's grep-the-defense-or-it-doesn't-exist rule caught it and kept the finding VALID, exactly where an unstructured review would likely have waved it through on the docstring's word.
- What is the Skeptical Triage SkillProof Score?
- 9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
- How do I install Skeptical Triage?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Skeptical Triage with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.