jobsearch-dashboard-ai

Use when reviewing, revising, testing, or gating public content drafts, LinkedIn posts, marketing copy, content-agent output, `/linkedin-content` behavior, prompt-fidelity reviews, publish readiness, or documentary build-in-public drafts for evidence, specificity, repetition, safety, grounded claims, realistic voice, and repairability.

Use whenever the user asks Codex to implement, execute, or ship a plan in this repo, especially requests that include saving to /plans, creating a branch, editing code, updating docs/wiki, running verification, committing, pushing, opening or updating a PR, and restarting dev. Captures the repo's required release workflow, safety boundaries, verification discipline, and PR standards.

Use when designing, redesigning, reviewing, or implementing dashboards, charts, command centers, analytics panels, review queues, decision surfaces, or other React UI in this repo where product usefulness, visual hierarchy, responsive layout, evidence visibility, action safety, accessibility, and screenshot verification matter.

Use when reviewing, mapping, designing, or changing Job Search OS architecture, especially when adding AgentType values, modifying Prisma models, changing workflow boundaries, adding routes/API surfaces, updating agent orchestration, revising ADK or LangGraph ownership, creating architecture reports, or documenting runtime source boundaries.

Use when creating, reviewing, or improving public build-in-public, LinkedIn, marketing, documentary, product narrative, or agent-content drafts for Job Search OS, especially `/linkedin-content` work that must be grounded in plans, agent runs, UI artifacts, analytics, screenshots, decisions, implementation notes, or privacy-safe source material instead of generic AI/product marketing copy.

Write or rewrite professional, staff-level GitHub pull request titles and descriptions for this repo. Use when creating a PR, updating a weak PR body, summarizing a branch for reviewers, preparing release notes from a diff, or converting implementation work into a contextual reviewer-ready PR with architecture, product, data, safety, verification, and rollout detail.