product
agentsview is for developers who run multiple AI coding agents and need to inspect, search, compare, and audit their local session history. They are usually in a debugging, review, or cost-monitoring workflow where dense, trustworthy data is more useful than promotional framing.
agentsview syncs agent session files into a local archive and serves a fast web UI for browsing sessions, searching transcripts, tracking usage, and reviewing costs across projects, models, and agents. Success means users can understand what happened across their agent runs without sending data to an external account or re-parsing raw files by hand.
Local-first, technical, focused. The interface should feel like an operational tool: direct, compact, and calm enough for repeated daily use.
Avoid marketing-site hero patterns, decorative SaaS dashboards, vague assistant-themed illustrations, and visual treatments that make local developer data feel like a hosted analytics product. Avoid hiding density behind oversized cards or ornamental motion.
- Put the session data first.
- Preserve local-first trust.
- Keep repeated workflows compact and predictable.
- Make freshness, filtering, and state legible.
- Prefer familiar product affordances over novelty.
No project-specific accessibility profile is documented yet. Default to clear focus states, keyboard-reachable controls, readable contrast in light and dark themes, reduced-motion-safe transitions, and chart labels that remain visible across supported viewport sizes.