alphabravo-oss

by AlphaBravo

Stops Claude from racing on ad-hoc work. Methodical-mode pre-response checklist injected every turn (Restate, Verified/Unverified assumptions, Read Floor, Approach Deliberation, Blast Radius, Stall Check, Competing Hypotheses, Goal-Driven Execution). Stop hook enforces four gates: file:line citation verifier, uncertainty-tell scanner, zero-tool-call grounding audit, and an iterative two-tier critic-dialog (Haiku fast-critic rounds 1-2, opus pre-stop-critic rounds 3-5, HARD block at round 5 until /adhoc:trust-me bypass). Each critic runs a 10-item rule compliance audit and sees the prior round's feedback to verify Claude actually addressed flagged issues. /adhoc:casual (one turn skip), /adhoc:trust-me (one-turn strict bypass), /adhoc:off (session disable). /adhoc:deep for heavier analysis, /adhoc:second-opinion for independent critique.

by AlphaBravo

Throw an AI dev at a problem and have it own the outcome. Five Opus agents (router, scout, worker, critic, fresh-eyes) collaborate to investigate, plan, execute, and verify — SSH into systems, spin up infra, edit code, ship features, debug deploys, get airgap bundles working. Worker self-switches between investigating, planning, executing, and verifying modes. Bounded scout budget, multi-agent reflexion on stuck (N=2-3 parallel variants + judge), critic and fresh-eyes must produce evidence ledgers of commands they actually ran. One checkpoint (plan approval) for side-effecting tasks, zero for pure investigation. Plan and debrief render in chat; state files exist in .crew/runs/ for resume. /crew:goal runs the worker→critic→fresh-eyes loop unattended until critic and fresh-eyes both return pass — enforced by a Stop hook, bounded by cycle and time caps.

by AlphaBravo

A mini-foundry — the build-verify-fix engine without the MCP server or the planning interview. You bring a structured spec (requirement IDs + acceptance criteria); crucible builds it and refuses to let it drift. /crucible:build <spec-file> decomposes the spec into units with DISJOINT file ownership, shows you the split for one approval, then runs a Workflow engine: CAST (parallel deliberate-engineering builders, each handed its unit prompt verbatim, each producing re-executable evidence bound to requirement IDs) -> INSPECT (blind orthogonal verification streams — wiring exists/substantive/wired, spec-before-code assay, and server-side re-execution of every evidence command) -> GRIND (every non-passing verdict is a defect; fix and full-re-inspect, loop until zero or a cycle cap) -> ASSAY (fresh-eyes agents read the spec first, form expectations, then confirm each requirement). Refuses a dirty tree; edits the shared working tree under disjoint ownership; commits per-unit only on your say-so so any unit is one revert away. Drops what needs the heavy machinery: no MCP gates/hashing, no isolated-worktree evidence, no brownfield flow-graph interview. Keeps the load-bearing principles: the engine routes and verifies, it never re-narrates the spec or judges the code itself.

by AlphaBravo

De-AI My UI — strips the tells that make an interface read as 'made by an AI in Tailwind.' Two modes, one shared catalog of ~19 slop signatures (font chaos, the purple-on-black AI palette, gratuitous parallax, weird hovers, default/neon gradients, neon gradient borders, recycled stock icons, everything-is-a-24px-card, over-rounded corners, huge drop shadows, the AI-startup serif, out-of-box shadcn, forced monochrome, icon/emoji overuse and mismatch, empty filler copy, tiny/low-contrast text). /damu:prevent renders the catalog's fixes into a copy-pasteable anti-slop ruleset, lightly tailored to the detected stack, for you to adjust and drop into CLAUDE.md or memory so Claude stops generating slop in the first place. /damu:remediate drives Playwright MCP to capture per-route screenshots plus extracted CSS facts (font counts and sizes, color clusters, WCAG contrast, radius/shadow distributions, gradient and parallax signals, shadcn/Tailwind-default signatures), then runs a holmes-style Workflow engine — a context pass that fixes what's legitimate for this product, nine blind per-tell lenses (typography, color, gradients & borders, shape & depth, layout soul, motion, iconography, copy, readability) that each hard-respect those legitimate patterns and self-rate confidence (HIGH only for unmotivated, uniform choices) so nothing is silently dropped, a completeness critic, and synthesis into a per-page deliberate/mixed/slop verdict with a ranked, file-anchored change list. Report first; --apply commits only HIGH-confidence low-risk fixes atomically and refuses a dirty tree. Governing principle: every tell is sometimes correct, so a finding is HIGH-confidence only when the choice looks unmotivated and uniform.

by AlphaBravo

End-to-end test authoring and maintenance for web apps via Playwright + Playwright MCP. Claude drives a real browser, reads the accessibility tree, and emits *.spec.ts files that pass — then keeps them passing. /e2e:init scaffolds Playwright in the cwd and registers the MCP server. /e2e:write <flow> drives the browser to perform the flow and emits a green spec. /e2e:crawl discovers every route and generates a smoke spec per route. /e2e:matrix produces a routes × roles coverage suite with reused storageState. /e2e:audit runs the suite, reads trace.zip files, and triages failures by root cause. /e2e:record wraps playwright codegen for manual capture. The author skill enforces selector hygiene (getByRole over CSS), auto-wait discipline, and idempotent test design; the diagnose skill drives trace-driven flaky-test triage. The spec-writer subagent isolates MCP-driven spec authoring from the main context.

by AlphaBravo

Codebase-aware specification engine for Claude Code. Runs parallel codebase research, then a grounded interview that captures implicit facts as A-AUTO-NNN entries, emits typed Global Invariants / State Transitions / Contracts tables, runs an adversarial spec review at R3.5, then writes a foundry-ready spec with version frontmatter (v2.1) — locked, classified, citation-backed. Detects brownfield, greenfield, and cosmetic modes; brownfield runs produce a flow delta against the existing system. Forge plans. Foundry builds.

by AlphaBravo

Autonomous build-verify-fix loop for Claude Code. Reads a Forge spec, decomposes it into castings with pre-authored teammate prompts, validates through an 11-dimension F0.9 gate (including F0.7 intent-carrier coverage), builds in parallel waves with server-side evidence re-execution and per-requirement-ID binding, runs up to 8 verification streams (TRACE, FLOW_TRACE, PROVE, RESEARCH_AUDIT, COVERAGE_DIFF, SIGHT, TEST/PROBE, TEST_OBSERVATIONS), grinds defects to zero, then assays with fresh eyes. Brownfield runs consume a flow delta; greenfield and cosmetic runs use the end-state pipeline. Forge plans. Foundry builds.

by AlphaBravo

Macro-architecture / design-cohesion review — the senior-engineer 'is this shaped right or accreted?' critique, not bug-hunting or lint. Reads a codebase like Holmes reads a room: deduces the real design problem from the detail that's out of place. /holmes:review <target> runs a Workflow engine that builds a grounded structural map of the target (a path, the diff, a subsystem, or a described piece of logic), then runs 7 blind design lenses — package/module proliferation, missed sharing & reuse, helper sprawl & abstraction fit, flow & layering, cohesion, naming/structural consistency, accretion markers — each a separate agent hunting one class of design problem. Every candidate finding is adversarially verified by a skeptic that tries to refute it as intentional, so plausible-but-wrong smells get dropped; a completeness critic then re-lenses any in-scope area no finding touched. The result is synthesized into a per-subsystem deliberate/mixed/accreted verdict with file:line evidence and an ordered, concrete reshaping plan. Report only — never edits code. Forbidden from style nits, correctness bugs, or treating comments as source of truth ('everybody lies'); an empty result is a valid honest outcome for a well-shaped target.

by AlphaBravo

Careful low-risk codebase cleanup across 7 focused tracks: deduplication, type consolidation, dead code removal, circular dependency untangling, type strengthening, error-handling cleanup, and deprecated/AI-slop removal. Each track inspects, writes a critical assessment, ranks proposed changes by confidence, and auto-applies ONLY HIGH-confidence LOW-risk changes per batch — running language-appropriate type checks, tests, and lints after each batch. MEDIUM and LOW confidence changes get surfaced for explicit human approval. Atomic commits per batch so any failing check is one git revert away. Designed for the conservative cleanup pass — never merge code that just LOOKS similar, never remove dynamically-imported code, never break legitimate boundary types, never silence real errors. Refuses to use comments as a source of truth.

by AlphaBravo

Always-on response shaping. A SessionStart hook loads a 10-rule ruleset into every new session — first line is the action, numbered multi-step work, restate where we are each turn, one concrete next action, concrete time estimates, visible wins, flat tone on failure, lists capped at 5, no preamble/recap/sign-off — plus six break-glass exceptions (depth requested, destructive action, debug spiral, real ambiguity, rule-would-delete-the-answer, harness outranks) and a pre-send delete pass. A UserPromptSubmit hook re-reads the state file every turn, so /tldr:off suspends the loaded ruleset immediately mid-session instead of next session. /tldr:verbose exempts a single turn and auto-reverts. /tldr:off, /tldr:on, /tldr:status, /tldr:help.

by AlphaBravo

Experiential UX review that drives the RUNNING app as a real user — not a code read. Reading source is a code review and structurally cannot find experience defects (a mode that doesn't do what its label claims, an overlay anchored to the wrong thing, a feature that silently breaks in one state, a table-stakes affordance that's simply absent). /ux-review:run launches the app, self-configures by discovering what the product is, who it's for, and what it promises, then builds a coverage grid (every screen × state × variant × viewport × input modality) and drives real jobs-to-be-done end-to-end through five sharp lenses — promise-vs-delivery, default-vs-buried, anchor/subject, every-state-not-the-happy-one, and what's-missing — plus the classic heuristics framed experientially. Adversarial states (empty/error/offline/slow/denied/first-run), a keyboard + screen-reader + mobile sweep, and console/network watching catch silent failures. A completeness critic loops until a full pass finds nothing new, and every finding is logged as job → expected → actual (+screenshot) → why-it-hurts → severity, ranked by user harm, with an auditable coverage report. Drives real browsers via Playwright MCP; the persona-driver subagent fans out one driver per persona × surface for exhaustive parallel coverage.

by AlphaBravo

Generate really good dynamic Workflow scripts on demand. /weave:make <task> classifies the task into an orchestration archetype (understand / design / review / research / migrate / generic), picks the right fan-out topology (pipeline by default, barrier only when a stage genuinely needs all prior results), and authors a self-contained Workflow script that obeys the Workflow tool's hard rules — pure-literal meta block, schema'd agent returns instead of text parsing, .filter(Boolean) after fan-out, adversarial-verify for findings, loop-until-dry / loop-until-budget for unknown-size work, isolation:'worktree' only for parallel file mutation, no Date.now/Math.random. Every script is validated with `node --check` and a 12-point self-audit before it is saved to .claude/workflows/ or run. Weave writes the orchestration so you don't freehand it.

by AlphaBravo

Production documentation for an app you built, in a fixed layout modelled on Harvester: subject-first directories and a Docusaurus site whose navigation is the filesystem. Every page declares both what it is and who it is for, with sections from The Good Docs Project and quality checks from the ISO/IEC/IEEE 26514 characteristics: type mixing, alt text, heading order, and reading grade measured against that page's own reader. Extracts the reference material, anchors every claim to a file and line, measures drift against a recorded manifest, and refuses to state what it could not check.