← ClaudeAtlas

really-honestlisted

Use this skill whenever Naveen wants Claude to re-examine its own previous response with genuine scrutiny. Triggers when Naveen types /really-honest, says "but for real", "are you actually sure", "what did you miss", "what's the weakest part of what you just said", or "be honest about this". This skill prevents the class of problems where Claude's first response is polished and complete-sounding but contains glossed-over assumptions, overconfident claims, or subtle errors. Apply after any high-stakes response, strategy output, plan, or technical recommendation. Sits alongside /honest in the Thinking & Reasoning shortcode category — but operates differently.
ElephantHunters/Navma-Skills · ★ 0 · AI & Automation · score 70
Install: claude install-skill ElephantHunters/Navma-Skills
# Really Honest A second-pass scrutiny skill. Forces Claude to break out of first-answer confidence and genuinely attack its own most recent output. Claude's default mode produces responses that sound complete and authoritative. This skill interrupts that pattern and demands a real second look. --- ## The Difference Between /honest and /really-honest | Shortcode | When | What it does | |---|---|---| | /honest | Before Claude responds | No sugarcoating — give raw truth upfront | | /really-honest | After Claude responds | Attack the response Claude just gave | /honest shapes the response before it's written. /really-honest scrutinises the response after it's written. They are complementary. Use /honest when you want truth from the start. Use /really-honest when Claude's output feels too clean or too confident. --- ## Claude's Behaviour When /really-honest Triggers Claude must re-read its previous response and answer these questions honestly: **What did I gloss over?** Was there a part of the answer where I moved on too quickly because I wasn't sure, or because going deeper would have complicated the response? **What assumption did I make that could be wrong?** Did I assume Naveen's situation matches the general case? Did I assume a tool or library behaves a certain way? Did I assume a timeline, a cost, or a difficulty level without knowing enough? **What is the weakest part of what I just said?** If someone were to challenge this response, where would they find the