explain
explain
Run uv run "$HERE/scripts/invoke.py" — $HERE is the directory that holds this SKILL.md — with everything the user typed after /explain, verbatim and however many lines, on stdin. Exit 0: do what it prints. On any other exit, if you introduced a known construction error and can correct it while preserving the user's request and authority, account for effects already produced, submit the corrected invocation through the same shim, and continue from the failed boundary; a refusal before the operation starts consumes no operation. Otherwise show what it printed to the user verbatim and stop. Never repair input the user supplied, or automatically retry exact help, an unmet dependency, an unrelated failure, or a failure whose origin or valid correction is unknown.
Arguments
If there is an argument, it is an instruction on what the user would like you to explain. The same instruction may arrive as a Contextual Instruction behind the reserved separator, so operands and instruction are read as one instruction between them. If no arguments are provided, the user wants you to explain your last response.
Steps
- Settle the subject: the reply just given, or what the instruction names in it or earlier in the conversation. Identify what needs clarification. If there is nothing to explain, say so and stop.
- Write the explanation for the reader below, under the instruction given — the language it names, the part it narrows to, the shape it constrains. Done when the reader's question is answered and any practical consequence or required decision is clear.
The reader
Write briefly and clearly to an intelligent, busy colleague; give just enough context to understand the answer. Assume ordinary subject knowledge but no familiarity with the project's glossary: use its terms consistently and briefly explain project-specific or unusual terms where needed.