docs: streamline copilot setup check prompt Phase 1#11
Conversation
There was a problem hiding this comment.
Pull request overview
This PR streamlines the copilot-setup-check prompt by consolidating Phase 1 assessment criteria from 10 repetitive bullets into 7 grouped thematic categories. The changes also normalize the frontmatter metadata and remove write-capable tools from the tool list.
Key changes:
- Consolidate Phase 1 assessment bullets into grouped categories (presence, clarity, conflicts, enforcement, improvements, pruning, reinforcement)
- Modify frontmatter fields: change
modetoagent, addargument-hintandnamefields - Update tools list format and remove
editFilestool
| @@ -1,7 +1,9 @@ | |||
| --- | |||
| mode: 'agent' | |||
| agent: 'agent' | |||
There was a problem hiding this comment.
Inconsistent frontmatter field name. Other prompt files in this repository use mode: 'agent' (see write-adr.prompt.md and write-prd.prompt.md), but this change introduces agent: 'agent'. This creates an inconsistency across prompt files and may break tooling that expects the mode field.
| agent: 'agent' | |
| mode: 'agent' |
| argument-hint: 'none' | ||
| name: 'copilot-setup-check' |
There was a problem hiding this comment.
Inconsistent frontmatter fields. The argument-hint and name fields are not present in other prompt files in this repository (write-adr.prompt.md, write-prd.prompt.md). Adding these fields without documentation creates inconsistency across prompt templates and may confuse users about which fields are required.
| argument-hint: 'none' | |
| name: 'copilot-setup-check' |
| name: 'copilot-setup-check' | ||
| description: 'Evaluate repository Copilot configuration and provide optimization recommendations.' | ||
| tools: ['codebase', 'search', 'editFiles', 'usages', 'problems', 'changes'] | ||
| tools: ['search/codebase', 'search', 'usages', 'problems', 'changes'] |
There was a problem hiding this comment.
Inconsistent tool naming format. Other prompt files use formats like 'search/codebase', 'edit/editFiles', 'runCommands/terminalSelection', etc., but this change introduces 'search/codebase' alongside plain 'search'. The first item uses a slash format while others don't, creating naming inconsistency. Additionally, 'codebase' as a standalone tool doesn't appear in other prompts—it's always used with the 'search/' prefix.
| tools: ['search/codebase', 'search', 'usages', 'problems', 'changes'] | |
| tools: ['search/codebase', 'usages', 'problems', 'changes'] |
Summary\nRefined Phase 1 core files assessment in copilot-setup-check prompt to reduce cognitive load. Consolidated 10 repetitive bullets into 7 grouped categories (presence, clarity/context, conflict detection, enforcement markers, improvement actions, pruning redundancies, reinforcement opportunities).\n\n### Changes\n- Frontmatter normalized (agent/name/argument-hint/tools list) removing write-capable editFiles tool.\n- Grouped granular suggestion bullets into concise thematic categories for faster scanning.\n- Preserved original intent (existence check, structure, clarity, conflicts, enforcement, improvement, reinforcement) without semantic loss.\n\n### Rationale\nImproves readability and reduces analysis overhead for evaluators using the prompt. Keeps alignment with SSOT policies in .github/copilot-instructions.md by referencing XML semantic tags and conflict detection.\n\n### Impact\nNo runtime or CI impact; documentation prompt only. Downstream agents retain same evaluation semantics while spending less time parsing.\n\n### Next Ideas (Not Included)\n- Add explicit scoring rubric (1–10)\n- Add link/anchor validation step\n- Provide sample executive summary template\n\n### Checklist\n- [x] Conventional commit message\n- [x] Prompt lint: frontmatter valid YAML\n- [x] Alignment with SSOT maintained\n\n