Prompt Analyze Skill

This skill runs only the execution-and-evaluation phase (Phase 1) of the prompt-builder skill. Use the prompt-builder skill's orchestration reference for the sandbox contract, the Prompt Tester and Prompt Evaluator dispatch matrix, and the cleanup contract. This skill adds only the analyze-only scope, the durable Analysis Report it writes under .copilot-tracking/reviews/logs/, and the Analysis Report structure in references/analysis-report-template.md.

Goal

Execute only Phase 1 of the prompt-builder skill for existing prompt artifacts: run the target prompts in a sandbox, evaluate them against the Prompt Design Principles and the Prompt Quality Criteria in prompt-builder.instructions.md, write a durable Analysis Report, and return markdown links to that report and to every evaluated artifact. This skill is read-only with respect to the analyzed artifacts: it never modifies the artifacts it analyzes, and its only writes are the sandbox logs and the durable Analysis Report.

Flow

1. Confirm the target prompt file(s) and derive the sandbox folder from the prompt-builder skill's sandbox contract in its orchestration reference. Reuse the same {{topic}} and {{run-number}} derivation to name the durable Analysis Report at .copilot-tracking/reviews/logs/{{YYYY-MM-DD}}/{{topic}}-{{run-number}}-analysis.md, where {{run-number}} matches the sandbox run number so repeated same-day, same-topic analyses never overwrite a prior report.
2. Dispatch Prompt Tester to execute the target prompt file(s) literally inside the sandbox and write an execution log, following the dispatch matrix in the prompt-builder skill's orchestration reference. When the only input is promptFiles, default the analysis purpose/requirements/expectations to "evaluate the target artifact(s) against the Prompt Quality Criteria."
3. Dispatch Prompt Evaluator to review the execution log and the target files against the Prompt Quality Criteria and write an evaluation log, following the same dispatch matrix.
4. Read the evaluation log and synthesize the Analysis Report from the evaluator findings using references/analysis-report-template.md. Write the report to the durable path from step 1, then present it inline as the final response. Stop after this phase and do not continue into research, build, or modification behavior.
5. Close the response with the Evaluated Artifacts and Report section near the end, listing a markdown link to each durable Analysis Report and to every evaluated artifact, following references/analysis-report-template.md.

Inputs

• promptFiles (optional): Existing prompt, instruction, agent, or skill artifact(s) to analyze. If omitted, use the current open or attached file(s).