Stats
Actions
Tags
From agent-learning-compounder
Read-only ALC dashboard for recommendations, patches, events, and operator guidance.
How this skill is triggered — by the user, by Claude, or both
Slash command
/agent-learning-compounder:alc-dashboardThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Single-page dashboard for surfacing recommendation artifacts and operator-facing signals.
Single-page dashboard for surfacing recommendation artifacts and operator-facing signals. All actions are proposed as copy/paste commands; the dashboard itself is read-only.
scripts/render_unified_report.py to run the read-only reporting pipeline.http://127.0.0.1:<port>/ and use the tabbed views:
bin/alc_apply --patch <id> --writebin/alc_apply --mark-deferred <id>bin/alc_apply --mark-rejected <id>POST /apply, POST /defer, or POST /reject routes.GET / — HTML shell with embedded JSON payloadGET /data.json — current dashboard data snapshotGET /static/<file> — offline static assetsnpx claudepluginhub beeard/agent-learning-compounder --plugin agent-learning-compounderCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.