melech-make-it-click

Solid

Recover when the agent's explanation did not land by finding the smallest thing blocking understanding and replacing the failed mental model with a concrete one.

AI & Automation 13 stars 3 forks Updated 2 days ago MIT

Install

View on GitHub

Quality Score: 82/100

Stars 20%
38
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
50
License 10%
100
Description 5%
100

Skill Content

# Make It Click The user invoked this because your previous explanation was smart, dense, or technically correct, but they still do not understand it. Do not defend, summarize, or merely shorten that explanation. **Find the smallest thing preventing the user from understanding. Explain that first using the most concrete route available. Do not simplify the entire previous answer; replace its failed mental model.** ## Recover the thread Use the conversation, especially the user's latest wording, to infer the exact sticking point. Often it is much smaller than the topic: - "Are we dropping the remaining data?" - "Where does this value exist while the program runs?" - "Why does this extra identifier exist at all?" Answer that question first, even if the previous answer covered ten other things. If the user's confusion is visible from context, do not ask them to identify it again. If two or more sticking points are equally plausible, ask one narrow question that distinguishes them instead of guessing. ## Change the route The previous route already failed. Choose a different one: - walk one concrete input through the system; - show a tiny value or code example; - state the before/after in plain language; - use an analogy only when it maps cleanly; - remove an unnecessary concept if complexity—not wording—is the problem. Prefer real nouns and actions over category names. Introduce one idea at a time. Define unavoidable jargon at the moment it appears. ## Response shape ...

Details

Author
AdirD
Repository
AdirD/agent-shell-hamelech
Created
1 months ago
Last Updated
2 days ago
Language
Python
License
MIT

Similar Skills

Semantically similar based on skill content — not just same category