sdlc-lite
claudecodexSDLC Lite agent team: leveled builders, planners, verifiers and more, dispatched as sub-agents on Claude and natively on Codex, at a fixed model and effort per level.
34 skills · 36 agents
Machine, repo and harness care: verification loops, flaky gates, port zombies, bare repos, macOS builds, cross-harness plugins, skill report cards.
Plugin dev-tools from marketplace sdlc-lite in kimgranlund/sdlc-lite. Commands read from the repo's manifest — 17 days ago.
author-cross-harness-pluginsCreates or migrates one-source plugins that load in both Claude Code and Codex, preserving skills, agents, commands, hooks, MCP wiring, scripts, and package metadata through shared cores and thin host adapters. Use when the user asks to build, port, package, audit, or repair a plugin for both harnes
closed-loop-runRun bounded, evidence-driven verification, diagnosis, repair, and learning cycles for any agentic task without collapsing into an unbounded recursive solver. Use when an agent must test or inspect work, identify failures, improvise locally through limited trial-and-error, progressively verify fixes,
flaky-gatesDecides whether a RED test gate can be trusted on a host running concurrent agents and builds: reap orphaned browser/compiler processes first, then separate contention from regression with red-under-load vs green-in-isolation. Use when the tests are flaky, red in CI but green locally, a different se
git-bare-repo-setupSet up, operate, and migrate to a bare-clone + sibling-worktrees local dev layout (git clone --bare, one worktree per branch, no privileged "main checkout"). Use when creating a bare layout for a repo, adding/removing branch worktrees, migrating an existing normal clone, or diagnosing tooling that b
macos-build-and-verifyBuild, package, and visually verify the native SwiftUI LG Controls Mac app, including its menu-bar interface and local Swift toolchain fallback. Use for this project's macOS builds and UI checks, not unrelated web apps or system-wide toolchain maintenance.
port-zombie-sweepFinds port-squatting dev servers (vite, next, node, workerd, python http.server, and similar) that outlived the session, worktree, or agent run that started them, and suggests which to kill, grouped by project. Use when a port is "already in use", a dev server picks a random port, --strictPort fails
skill-report-cardDiscover local skill artifacts and evaluation evidence, then produce an evidence-backed KPI dashboard. Use for skill performance reporting, goals, measurement coverage and trends; not to grade a skill from its instructions alone.
SDLC Lite agent team: leveled builders, planners, verifiers and more, dispatched as sub-agents on Claude and natively on Codex, at a fixed model and effort per level.
34 skills · 36 agents
UI construction, Differential Forms geometry, HCT palettes, inspector jigs, component variants, wireframe review and design critique.
8 skills
Progressive disclosure, capability contracts, evidence, bounded reuse, context-routing diagnosis, system atlases and ICM workspaces.
6 skills
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.