Ralf Loop
>-
Parallel Refactor
>
Repo Structure Audit
Define and audit repo structure for code and knowledge projects. Use when starting a repo, moving a project into a standard layout, editing AGENTS.md or agent settings, checking root-file sprawl, auditing projects under Code, or asking whether a repo is organized correctly.
Cyborg Check
Audit whether a judgment-encoding system closes the loop between captured patterns and real outcomes, or is becoming an org-scale cyborg that spreads unverified judgment at machine speed. Trigger on "/cyborg-check", "cyborg check", "does this close the outcome loop", "is this just captured judgment", "capture without grading", "judgment audit", "are these patterns verified", or when assessing an…
Runtime Guard
Prevents an agent from editing or publishing without clearly defined authority, boundaries, stop conditions, or rollback.
60 30 10
>
Humanizer
|
Autoresearch
Eval-driven keep/discard optimization loop for any laptop-runnable target — Open Brain retrieval, a skill's routing/description, a prompt or talk-track, an email template, a deck outline, a config, any editable surface you can score against a frozen labeled test set. Measure a baseline, make ONE change, re-measure, keep if the metric improved, revert if not, log every experiment, repeat until it…
Context Window Audit
Audit an agent setup for token waste, context-window bloat, stale instructions, and dead command wrappers. Use when the user says context-window audit, context audit, audit my context, check my settings, why is Codex so slow, token optimization, or asks why startup context is too large.
Truth Layer Os
MCP-first truth-layer harness for defensible AI-generated Office artifacts
Skill Tune
Audit or refactor a skill or prompt artifact for prompt technical debt. Use when explicitly asked to review SKILL.md, AGENTS.md, CLAUDE.md, system prompts, command wrappers, or tool instructions for prompt debt, over-steering, context bloat, stale model-specific instructions, duplicated canonical rules, misplaced executable checks, or missing authority boundaries. Use as a debt review pass before…
N Agentic Harnesses
>-
Quality Check Eval
Sort AI output quality checks into four buckets: automate in code, judge with an LLM, keep for human review, or remove. Use when reviews are too manual, evals are vague, a workflow needs clearer quality gates, or the user wants to decide which AI output checks are worth automating. Not for checking whether a system grades encoded judgment against real outcomes; use cyborg-check for that.
Swarm
Invocation-only workflow. Use only when the user includes literal `swarm` as an instruction to apply this workflow. Do not trigger for broader agent, harness, judge, checker, orchestration, or multi-agent requests unless the literal `swarm` invocation is present.
Skill Creator V3
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
Weekly Signal Diff Ai
|
Research Synthesis
|
Run Kit
Design, route, audit, or cross-check AI agent runs. Use when the user wants to turn a fuzzy task into an agent-ready assignment, decide whether to steer or dispatch work, inspect whether an agent result is real, cross-check one agent's output with another, package agent-run prompts, or talk about proof, review burden, source of truth, permissions, completion theater, or understanding theater.