caveman
claudeTalk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Evidence-backed five-phase investigation for Access/VBA and SQL Server legacy applications, plus an optional Django REST + React modernization pipeline that carries a project from that investigation's output to implemented, coverage-traced code.
Plugin ak from marketplace access-modernization-kit in tuanvtanrakutei/access-modernization-kit. Commands read from the repo's manifest — 29 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
bootstrap-projectSet up a brand-new modernization project - copy the template documents into a target repo's docs directory, and seed Screens_Registry.md from a five-phase run's Phase 2 inventory when one exists. Trigger when the user wants to start a new Access-family modernization project, bootstrap a project, or
code-screenImplement one screen from its existing, already-frozen screen plan - backend then frontend (Stages 3a-3b). Refuses to start unless Stages 1-2 already produced a two-contract screen plan with a populated gap matrix. Trigger when the user explicitly wants just the coding stage for a screen that alread
dev-stackBring a modernization project's Docker stack up, check it, or stop it - backend, frontend and a local development database, with migrations applied. Trigger when the user wants to start, run, check or stop the project locally, or a check needs the stack. Examples: \"/ak:dev-stack up\", \"/ak:dev-sta
investigateAnalyze Microsoft Access VBA applications and MDB/ACCDB/ADP projects connected to SQL Server through a mandatory five-phase, evidence-backed legacy-system investigation with Access extraction, deterministic module planning, and provider-neutral multi-agent orchestration. Use when an agent must packa
modernize-screenRun the end-to-end legacy Access modernization pipeline for one screen or several in parallel — computed scope, screen plan, backend coding, frontend coding, tests, and review. Trigger when the user wants to implement, refresh, test, or review a screen of a legacy Access application being rebuilt as
plan-screenRun Stages 1 and 2 only for one screen - the computed scope and the screen plan - without touching code. Trigger when the user explicitly wants just the planning and documentation stages for a screen, not full implementation - modernize-screen already covers straight through implementation, so use t
review-screenReview one screen against every upstream artifact and record a verdict (Stage 5). Reviews whatever upstream artifacts exist and says what's missing, rather than refusing outright. Trigger when the user explicitly wants a review or a verdict for one screen. Examples: \"/ak:review-screen OrderEntry\",
screen-statusRead-only status for one screen or the whole registry - artifacts present, track status, open findings. Writes nothing, always safe to run. Trigger when the user wants to check where a screen or the whole project stands without starting any work. Examples: \"/ak:screen-status OrderEntry\", \"screen-
test-screenWrite and run the tests for one screen - backend suite then frontend end-to-end (Stages 4a-4b) - and close the stage with the evidence and gate G4: which rules the tests back, whether they can fail, whether the output matches the legacy output. Refuses to start unless a coding record already exists
triage-suiteWork out what is actually wrong with a failing test suite by grouping failures by error signature instead of by file, so a wall of red resolves into a few root causes. Trigger when a suite is red and the cause is unclear, when there are more failures than anyone wants to read one at a time, before d
validate-docsCheck a bootstrapped modernization project's documentation set for defects a reader will not notice — unfilled config keys, unresolved placeholders, dangling document references, malformed or duplicated issue rows, issue numbers cited by code with no row, resolved rows citing files that no longer ex
python "${CLAUDE_PLUGIN_ROOT}/hooks/scope_sensor.py"
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.