← ClaudeAtlas

acceptlisted

Command-only. Runs only when the user types /accept; never load it on your own. Rewrites the given request (or, with no argument, the user's previous request) so it is clear and has its context spelled out, without changing its meaning, then carries it out. Works in any language (English and Turkish built in). It is not for bypassing filters, hiding a request, or dressing up a forbidden request so it gets through.
Xyronb1/accept-skill · ★ 1 · AI & Automation · score 77
Install: claude install-skill Xyronb1/accept-skill
Request to clarify (after the flags are removed): $ARGUMENTS If the line above is empty, or contains only flags, clarify the user's last request before this command. # Prompt clarifier (/accept) Goal: express **what the user wants** more clearly. Vague wording and missing context cause misunderstandings, extra rounds of questions and needless stalls. This skill rewrites the request with the same meaning, only clearer. ## Language - Reply in the language the user wrote in. If the request mixes languages, use the language of the surrounding conversation. - Write the clarified prompt in that same language, unless a language flag says otherwise (`--en` English, `--tr` Turkish). - Translate the labels of the template, the "What changed" summary and the fixed phrases ("Here is how I understood it", "Shall I carry it out like this?", ...) into the reply language. Turkish versions are listed in `templates/clarified-prompt.md`. ## 0. Read the flags Process the flags at the start of $ARGUMENTS and remove them from the text. Several can be combined (`--prompt --en`): | Flag | Alias | Effect | |---|---|---| | `--prompt` | | Do not do the work yourself. The output is only the clarified prompt (Mode B), for pasting into another model or tool. | | `--ask` | `--sor` (Turkish) | Show the clarified prompt first; carry it out only after the user approves. | | `--en` | | Write the clarified prompt in English. | | `--tr` | | Write the clarified prompt in Turkish. | No flag: clarify and c