القائمة
Use when designing ShotgunCV AI feedback, correction, override, retry, reset, preference, or human-in-the-loop controls.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Use when designing ShotgunCV AI output, recommendations, scoring, ranking, summaries, generated resumes, or model-assisted decisions.
Use when designing ShotgunCV tables, ranking lists, filters, sorting, bulk actions, comparison views, or data-dense review queues.
Use when defining ShotgunCV frontend UI foundations, design tokens, component states, accessibility rules, layout density, or implementation handoff constraints.
Use when designing ShotgunCV dashboards, operational workspaces, run viewers, detail pages, monitoring pages, or workflow-heavy product screens.
| name | designing-ai-feedback-control |
| description | Use when designing ShotgunCV AI feedback, correction, override, retry, reset, preference, or human-in-the-loop controls. |
ShotgunCV feedback controls should keep the user in charge of local artifacts and repeated job-application decisions. Feedback improves the current run workflow first; it should not imply model training, remote sync, or hidden automation.
| Feedback type | Best UI |
|---|---|
| Fast quality signal | Useful/not useful, approve/reject, compact reason menu. |
| Correction | Inline edit, field-level correction, replacement text, evidence override note. |
| Workflow decision | Accept recommendation, skip, defer, compare alternatives. |
| Local preference | Remember locally, stop suggesting this, reset preference. |
| Failure report | Wrong source, missing evidence, hallucinated claim, unsupported rewrite, unsafe suggestion. |
run_dir artifacts without an explicit confirmation path.