Jason Hall
Community@imjasonh · Brooklyn, NY
Doin stuff
Agent Skills by Jason Hall
Showing 38 vetted skills indexed across 1 GitHub repositories.
imjasonhfigure-it-out
Designs an auditable phased playbook for large migrations and ambiguous multi-part changes.
imjasonhbro
Restates the previous AI message in plain language without jargon.
imjasonhinterrogate
Spawns multiple LLM reviewers to adversarially review code changes and synthesizes a verdict.
imjasonhprinciple-boundary-discipline
Concentrate validation and error handling at system boundaries while keeping business logic pure.
imjasonhapple-hig
Applies Apple Human Interface Guidelines when designing and reviewing SwiftUI and UIKit interfaces.
imjasonhtechnical-writing
Applies Diátaxis, Google style, STE, and Global English rules to technical documentation.
imjasonhprinciple-fix-root-causes
Guides debugging by tracing symptoms to root causes instead of applying symptom fixes.
imjasonhprinciple-encode-lessons-in-structure
Convert recurring corrections into lint rules, metadata flags, runtime checks, or scripts.
imjasonhhow
Generates architectural explanations of codebases by orchestrating parallel read-only explorer and explainer subagents.
imjasonhblast-radius
Analyze what a code change could break beyond the diff and prove safety by running real code.
imjasonhprinciple-subtract-before-you-add
Guides removal of dead code and redundant logic before building additions or refactors.
imjasonhreflect
Spawns three parallel review subagents over a session transcript to route learnings into skill edits.
imjasonhprinciple-foundational-thinking
Guides structural decisions on data structures, scaffolding order, and concurrency before writing logic.
imjasonhshow-me-your-work
Maintains an append-only TSV decision log with evidence pointers for long-running or unattended work.
imjasonhprinciple-build-the-lever
Builds rerunnable codemods, scripts, and generators instead of performing edits by hand.
imjasonhPoteto Mode
Orchestrates disciplined agent workflows with playbooks, subagents, and verified engineering principles.
imjasonhprinciple-make-operations-idempotent
Designs state-mutating operations to converge to the same end state across crashes, restarts, and retries.
imjasonhprinciple-test-behavior-not-implementation
Rewrites or deletes tests that assert implementation details instead of observable behavior.
imjasonhgoogle-developer-style
Applies Google developer documentation style rules to comments, docs, and user-facing prose.
imjasonhMake Bot UI
Builds a local web UI that wakes a Grok Bot through a webhook routine over Tailscale.
imjasonhprinciple-outcome-oriented-execution
Guides planned rewrites and migrations toward target architecture with explicit verification boundaries.
imjasonharchitect
Designs software architecture through parallel multi-model sketching before implementation begins.
imjasonhprinciple-never-block-on-the-human
Guides agents to proceed on reversible work instead of pausing for human confirmation.
imjasonhprinciple-redesign-from-first-principles
Redesigns existing code as if new requirements were foundational assumptions from the start.