updraft

Plugin updraft in windroc-labs/updraft. Commands read from the repo's manifest — today.

Skills (31)

• council

  Use at a genuine fork, a choice between viable options with materially different trade-offs or one costly to reverse, to get independent picks: 2–4 fresh subagents get one neutral brief (the question and options, the author's own words and the documents they name, read-only code) and each returns a

• grilling

  Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

• light-build

  Low-ceremony autonomous build: work on the current feature branch or a new isolated worktree, write a task list, implement each task itself, and run a capped review by every core lens plus the optional lenses the change matches to a single review-ready branch, plus a draft PR when the repo has CI or

• principle-attack-the-premise

  Apply when two or more fixes that share one premise have failed the same gate. Write the premise down and test it directly before you write another fix that assumes it.

• principle-boundary-discipline

  Apply when wiring validation, error handling, or framework adapters. Concentrate guards at system boundaries (CLI, config, network, external APIs); trust internal types and keep business logic in pure functions.

• principle-build-the-lever

  Apply to any non-trivial work, not just bulk work: edits, migrations, analyses, checks. Build the tool that does it or proves it (codemod, script, generator, or a skill your subagents follow) instead of working by hand. The tool is the artifact a reviewer can rerun.

• principle-encode-lessons-in-structure

  Apply when you catch yourself writing the same instruction a second time, or notice a recurring correction. Encode the rule as a lint, metadata flag, runtime check, or script instead of more text.

• principle-exhaust-the-design-space

  Apply when facing a novel UI interaction or architectural decision with no precedent in the codebase. Build 2-3 competing prototypes and compare side by side before committing.

• principle-experience-first

  Apply when product, UX, or feature-scope tradeoffs come up. Choose user delight over implementation convenience; ship fewer polished features over more rough ones.

• principle-explain-the-number

  Apply before you measure, and before you trust, report, or act on a number you measured: a speedup, a regression, a throughput, a latency, or an eval result. Find what limits it, and rule out that it measured something other than the work you think.

• principle-fix-root-causes

  Apply when debugging. Trace each symptom back to the first place the system went wrong and fix it there. A guard is a fix only when the state it handles is legitimate.

• principle-foundational-thinking

  Apply before writing logic: choosing core types and data structures, sequencing scaffold-vs-feature work, asking what concurrent actors share. Get the data structures right so downstream code becomes obvious.

• principle-guard-the-context-window

  Apply when context is filling up: large outputs, long files, repeated reads, fan-out planning. Route bulk to subagents; keep summaries in the main thread, not raw payloads.

• principle-laziness-protocol

  Apply when refactoring, evaluating diff size, or tempted to add abstractions, layers, or signal threading. Bias toward deletion and the smallest change that solves the problem.

• principle-make-operations-idempotent

  Apply when designing commands, lifecycle steps, or processing loops that run amid crashes, restarts, and retries. Converge to the same end state regardless of partial prior runs.

• principle-migrate-callers-then-delete-legacy-apis

  Apply when introducing a new internal API while old callers still exist. Migrate callers and delete the old API in the same wave instead of preserving compatibility layers.

• principle-minimize-reader-load

  Apply when reviewing or shaping code that's hard to trace. Count layers between question and answer, and hidden state in the reader's head; collapse one-caller wrappers and shrink mutable scope.

• principle-model-the-domain

  Apply when writing stateful logic, or when code branches a lot or repeats a shape assumption across files.

• principle-never-block-on-the-human

  Apply when tempted to ask 'should I do X?' on reversible work.

• principle-outcome-oriented-execution

  Apply during planned rewrites and migrations with explicit phase boundaries. Converge on the target architecture; don't preserve smooth intermediate states with throwaway compatibility code.

• principle-prove-it-works

  Apply after completing a task, before declaring done. Verify against the real artifact (run the feature, read the actual value, inspect the diff), not a proxy, self-report, or 'it compiles.'

• principle-redesign-from-first-principles

  Apply when integrating a new requirement into an existing design.

• principle-separate-before-serializing-shared-state

  Apply when concurrent actors might write to the same file, branch, key, or state object. Eliminate the sharing first; serialize structurally only when one shared writer is a real invariant.

• principle-sequence-verifiable-units

  Apply to multi-step work (sweeps, migrations, runs of similar edits) and to how you stack commits and PRs. Break work into small units that each end in a verifiable state, check each before the next, and order delivery so the sequence proves itself to a reviewer.

• principle-subtract-before-you-add

  Apply when sequencing an addition, refactor, or rewrite. Remove dead code, redundant validators, and stub references first, then build on the simpler base.

• principle-test-behavior-not-implementation

  Apply when you write, change, or keep a test. Assert what the code's users can observe, so that the test fails when that behavior breaks and keeps passing when only the internals change.

• principle-type-system-discipline

  Apply when designing types, reviewing a function signature, or writing code in any statically-typed language. Make illegal states unrepresentable, brand semantic primitives, parse external data at boundaries, exhaust variants, derive from authoritative schemas.

• principles

  Use to find and read, by name, one of pstack's 24 engineering principles (such as fix-root-causes or prove-it-works) or its unslop writing method, bundled in this plugin. Lists each name with when it applies and its file.

• review

  Use to review a document (a spec, PRD, design or plan) against its author's own words with independent reviewer agents, fixing it yourself until every reviewer passes or the round cap. Works without a git repo, with this plugin's reviewers or a folder of your own. Pass doc=<path> requirement=<text o

• to-tickets

  Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker (edges as text in one file per ticket locally, or native blocking links on a real tracker).

• unslop

  Cut AI tells from any writing.

Subagents (11)

• architecture-reviewer

  Dual-phase architecture reviewer (read-only): component boundaries, coupling and cohesion, extensibility, interface clarity; the spec's design in the spec phase, the produced structure in the work phase. Returns the strict verdict grammar.

• code-quality-reviewer

  Optional work-phase code-quality reviewer (read-only, code files): readability, naming, duplication, dead code, needless complexity, and whether comments earn their place. Returns the strict verdict grammar.

• correctness-reviewer

  Work-phase correctness reviewer (read-only): logic errors, edge and boundary cases, error paths and failure modes. Returns the strict verdict grammar.

• doc-reviewer

  Work-phase documentation reviewer (read-only): finds docs anywhere in the repo that the change made false, by a bounded, change-anchored search, and flags bloated doc edits. Returns the strict verdict grammar.

• performance-reviewer

  Optional work-phase performance reviewer (read-only, code files): algorithmic complexity, queries or expensive calls in loops, needless allocation, unbounded growth, resource leaks, hotspots. Returns the strict verdict grammar.

• requirement-fidelity-reviewer

  Work-phase requirement-fidelity reviewer (read-only): does the work deliver the original requirement, traced through the spec, with nothing dropped, drifted or added? Tags each item and returns the strict verdict grammar.

• reviewer-contract

  Authoring template, not a dispatchable reviewer. Its body is the single source of the contract and verdict grammar copied into ordinary reviewers; the verification reviewer has a dedicated execution contract. Selector-inert: no phase, tier, lens or applies_to.

• security-reviewer

  Optional dual-phase security reviewer (read-only): in the spec phase, whether the spec states the right security requirements; in the work phase, whether the work upholds them. Authz, input validation, secrets, injection, supply chain. Returns the strict verdict grammar.

• spec-fitness-reviewer

  Spec-phase spec-fitness reviewer (read-only): does the spec satisfy the requirement, with no gaps, ambiguity or scope error, and is it testable? Returns the strict verdict grammar.

• test-reviewer

  Optional work-phase test reviewer (read-only, code or test files): tests exist for new and changed behavior, assert the spec'd results, cover edges, and would fail if the code broke. Returns the strict verdict grammar.

• verification-reviewer

  Work-phase verification reviewer. Runs the complete target-repo checks in the frozen source worktree at a fixed revision and returns exact evidence.

Similar plugins