Score past claims against actual outcomes to evaluate decision accuracy.
npx skills add https://github.com/ronniepinnell/casper --skill calibrate-ronniepinnell
Please help me install this Agent Skill. Skill: calibrate Source: https://github.com/ronniepinnell/casper/tree/main/skills/calibrate Command: npx skills add https://github.com/ronniepinnell/casper --skill calibrate-ronniepinnell
This Skill assesses the accuracy of past decisions, helping to refine confidence levels and prevent future mistakes.
Run /calibrate last milestone to score the accuracy of the past decisions made in the last milestone.
High-intent search queries and answers about installing and using this skill.
Assess the accuracy of historical project decisions by scoring past claims against actual outcomes using project logs, providing feedback to refine confidence levels for future risk management.
Decision calibration in project planning is the process of reviewing and rating past risk assessments and estimates against actual results to understand how well they predicted outcomes.
Review past risk assessments after a project milestone by running a calibration analysis on your historical data, which scores past claims to reveal insights into your decision accuracy.
Yes, you need access to project logs and historical data to analyze past decision accuracy, as this historical information is required to score claims against actual outcomes.
Analyzing past claims improves future decision-making by identifying patterns in prediction accuracy, allowing for better resource allocation and more informed planning in subsequent project phases.
The best way to score past claims for correctness is to systematically compare documented historical estimates against actual milestone outcomes, generating an accuracy evaluation that refines confidence levels.
ARCASSystemsAudit project memory claims against operational files and suggest resolutions.
k97Decompose project objectives into time-estimated tasks with dependencies and milestones.
alfredxwOrchestrate multi-milestone project workflows across threads and subagents.
EnesMeyzin98Compare predicted confidence bands to realized outcomes to flag miscalibration.
