Newton-School

Start and build software projects through a local architecture review gate. Use when a user invokes AgentFlow, asks to initialize a project with AgentFlow, or works in a repository that already contains `.agentflow/`.

Create or revise accurate AgentFlow table, flow, and document pages from requirements, code, and review feedback before implementation.

Create or revise precise app-owned Django architecture using AgentFlow tables, endpoint and task flows, and minimal platform documentation.

Compare a client AgentFlow harness with the latest published release and safely apply compatible updates when the user explicitly requests it.

Create or revise accurate AgentFlow table, flow, and document pages from requirements, code, and review feedback before implementation.

Create or revise precise app-owned Django architecture using AgentFlow tables, endpoint and task flows, and minimal platform documentation.

Apply open review comments to a Django AgentFlow proposal, revalidate app ownership and related contracts, and preserve immutable review history.

Convert approved Django AgentFlow architecture into an exact app-owned implementation and migration plan without adding unapproved scope.

Design DSDS chaos experiments with deterministic setup, explicit hypotheses, and measurable outcomes. Use when planning failure drills, resilience validation, pre-production risk exercises, and repeatable chaos scenarios for regression testing.

Map architecture requirements to valid DSDS component categories and component types. Use when converting system requirements into simulator node selections, resolving ambiguous type choices, and ensuring required attributes are captured before topology…

Map DSDS architectures across AWS, GCP, and Azure equivalents and estimate cost-impact tradeoffs. Use when planning provider portability, comparing cloud options, and explaining cost/performance implications of design choices.

Convert DSDS UI specs, mock HTML screens, and system docs into concrete frontend component implementation tasks. Use when turning design references into React components, hooks, state contracts, and acceptance criteria.

Analyze likely cascade paths and propagation semantics in DSDS architectures. Use when estimating blast radius, diagnosing multi-hop degradation, validating containment strategies, or explaining how a localized failure can become systemic.

Define and evaluate deterministic invariants and policy checks for DSDS simulations. Use when encoding correctness rules, post-run validation gates, compliance constraints, and safety checks that must hold across scenarios.

Tune DSDS resilience controls including retries, deadlines, circuit breakers, rate limits, bulkheads, and load shedding. Use when balancing availability, latency, and resource protection under failure and saturation conditions.

Compose deterministic DSDS simulation scenarios from resilience hypotheses, load-test goals, or failure drills. Use when defining workload plus fault plus invariant bundles, built-in presets, and chaos experiments with reproducible seeds and explicit…

Convert a Framer-published website into a fully self-contained, pixel-perfect coded static site that depends on nothing from Framer's servers. Use whenever a user pastes a Framer site URL (e.g. *.framer.website, *.framer.app, or a custom domain built on…

Hybrid QA investigation for web applications. Use when a coding agent needs to test a feature or flow, reproduce or triage a browser bug, validate regressions, inspect failures with Playwright or browser diagnostics, or generate a QA report with scenario…

Convert an SVG into a working Lottie/Bodymovin JSON animation based on a plain-English brief. Use this skill whenever the user gives you an SVG (paste, file, or inline markup) plus any description of how it should move, OR mentions "Lottie", ".lottie",…

Root-cause analysis for a Watchtower / miniOG job using its job id. Use when the user says "check what happened to job", "RCA this job", "why did this job fail", "investigate job id", or provides a job UUID and asks what went wrong. Queries the local SQLite…

Use this skill whenever the user wants to create, write, or add a case study for Koyo's AI Interviewer / AI Recruiter product. Triggers include: any mention of "case study", "customer story", "testimonial write-up", "success story", or requests to document…