Austin Theriot
Community@austintheriot · Austin, TX
Agent Skills by Austin Theriot
Showing 21 vetted skills indexed across 1 GitHub repositories.
expert-review
Classify code regions and dispatch specialist evaluators to produce ranked review reports.
agent-modes
Standardize output formats for review, plan, and consult agent workflows.
expert-consult
Dispatch 2–4 expert agents in parallel and synthesize disagreement into a structured summary.
fp-design
Guide functional-programming design decisions by brainstorming or critiquing designs.
oo-review
Identify object-oriented patterns in code changes and evaluate their implementation.
data-flow-review
Review code diffs for data-flow, state-ownership, and boundary mismatches.
distsys-review
Review distributed-systems code diffs for correctness and operability hazards.
observability-design
Generate structured instrumentation, SLO, and alerting design documents.
comms-and-team
Draft or critique team communication artifacts using people-and-org frameworks.
rust-review
Review Rust diffs for idiomatic design, safety, and footguns.
fp-review
Analyzes code diffs and reports severity-labeled FP improvement opportunities with file:line references.
ts-review
Review TypeScript diffs for type-design and type-safety issues.
product-design
Generate product briefs or severity-based critiques of product plans.
analytics-design
Design Mixpanel-aware tracking plans or critiques for web analytics instrumentation.
expert-plan
Interview users to produce an auditable Markdown feature specification.
system-design
Guide system design brainstorming and critique with tradeoffs and failure-mode analysis.
observability-review
Review code changes for observability quality and OpenTelemetry compliance.
consult
Route user questions to a single consult-capable subagent for direct expert advice.
pr-diff
Generate a three-dot diff with merge-base logic and structured Markdown summaries.
oo-design
Brainstorm or critique object-oriented architectures and type models.
one-by-one
Guide sequential item-by-item decisions on lists with end-only synthesis.