Interpreting SLOPE handicap cards, miss patterns, hazard data, and generating actionable recommendations. Use when analyzing sprint performance, reading handicap cards, investigating miss patterns, computing dispersion, or generating backlog strategies from performance data.
Sprint lifecycle orchestration for SLOPE-managed projects. Use when starting sprints, scoring tickets, building scorecards, running reviews, or following the routine hierarchy. Use when user says "sprint", "scorecard", "pre-round", "post-hole", "briefing", "review", "hazard", or "commit discipline".
Institutional knowledge for the Slope autonomous development loop. Use when executing automated sprints, reviewing loop results, analyzing model tier performance, or working on Slope's own codebase via the loop. Use when user says "run a sprint", "check loop status", "review sprint results", "model performance"…
Run the session-close routine. A session is not a sprint: a long session can span several sprints or end mid-sprint, so this complements /post-sprint rather than replacing it. Leaked claims and unpushed work are the two failure modes this prevents.
Respond to SLOPE guard warnings in Codex without bypassing sprint discipline. Trigger when a SLOPE guard warns, blocks, recommends a review, asks for a claim, or reports missing sprint context.
Run SLOPE post-merge retros from Codex and persist durable learnings. Trigger after a pull request merges, when the user asks for a retro, or when sprint closeout should save lessons into memory.
Set up SLOPE in a Codex-managed repository without requiring the user to know CLI details. Trigger when the user asks to install, initialize, update, or verify SLOPE for Codex.
Use SLOPE from Codex to inspect, start, claim, or close a sprint in a repository with .slope state. Trigger when the user asks to set up SLOPE, register sprint work, check active sprint status, or close out sprint work.