Score agent kernel calibration metrics against billed usage data.
npx skills add https://github.com/anthony-chaudhary/fak --skill dojo-rsi-score
Please help me install this Agent Skill. Skill: dojo-rsi-score Source: https://github.com/anthony-chaudhary/fak/tree/main/.claude/skills/dojo-rsi-score Command: npx skills add https://github.com/anthony-chaudhary/fak --skill dojo-rsi-score
This skill addresses the risk of drift in agent calibration by ensuring that the dojo's self-improving RSI loop remains anchored to real, billed usage rather than fabricated metrics or plan-mode scaffolds.
Run the dojo-rsi-score skill to evaluate the current calibration report and propose necessary recalibrations based on your billed usage.
High-intent search queries and answers about installing and using this skill.
To validate agent kernel calibration, you must score performance metrics against a valid billed corpus report. This ensures recalibrations are grounded in empirical evidence and protected from floor-breach errors.
The dojo RSI framework is a calibration loop for agent kernels that prevents metric drift. It anchors self-improvement to genuine billed usage data rather than fabricated metrics or plan-mode scaffolds.
Yes, the fak binary is required to execute the fold, propose, and score sequence. You also need a valid billed corpus report to run the dojo RSI calibration workflow.
You should run an RSI calibration loop after updating a dojo lever or performing a fresh corpus scan. This ensures the gym's predictors remain accurately calibrated against real-world performance.
Dojo calibration fails without a valid billed corpus report to provide genuine FoldCalibrable drops. The automated scoring sequence requires empirical data to rank, preview, and self-score recalibrations.
mkalkereReproduce, isolate, and diagnose code failures with evidence-based techniques.
hiddink-aiAudit and optimize AI agent settings to minimize token consumption.
lucking7Dispatch a block-scalar-tools agent to scan a corpus.
Jason5330Automate rubric and bucket boundary upgrades with auditable governance.
