caveman
claudeTalk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
The agent's workflow intelligence: the /workflows:add-item, /workflows:author, /workflows:run-item, and /workflows:discuss commands plus the workflows MCP server — record, route, author, discuss, and drive workflow work from Claude Code. The floor to use the product; no tray required.
Plugin workflows from marketplace workflows in codingawayy/workflows-releases. Commands read from the repo's manifest — 14 days ago.
This plugin runs code on your machine — it starts local MCP servers. The exact commands are listed below; read them before installing.
workflows.add-itemUse when capturing work in the tracker, breaking it into items, routing each item to a workflow, or recording dependencies and epics.
workflows.authorUse when creating or changing a workflow definition, including its transitions, steps, documents, conditions, outcomes, or input contract.
workflows.build-skillsInspect every workflow in the connected project and propose, create, reconcile, or remove project-local agent operations for named transitions and human intervention. Use when the user wants workflow-specific skills, slash commands, or an operations surface generated from the project's live workflow
workflows.discussUse when discussing, understanding, questioning, or amending one workflow item in conversation.
workflows.run-itemDrive a workflow item through its autonomous transitions from THIS interactive session — the next_step → execute → submit_step loop — instead of the headless engine. Invoke when the user wants to run, drive, advance, or work a backlog item in-session (e.g. "/workflows:run-item B26.070", "drive B26.0
workflows.unblockUse when inspecting a project's human-intervention queue, reviewing selected waiting items, guiding the user through one intervention at a time, or interviewing the user through the decisions of many waiting items in one batch review.
workflows.writeThe system's writing-craft guidance for any document, report, or written deliverable — lead with the conclusion, orient before detail, visualize, write for the cold-start reader who has only the documents, and be concise. Invoke when composing or tightening written content, especially outside a work
/workflows:add-itemTurn natural-language work into well-recorded items — break it down, route each piece to the right workflow, record it to that workflow's quality bar, and (on approval) create the items with their dependencies and epics.
/workflows:authorBuild a new workflow or amend an existing one to good shape — its transitions, step prompts, documents, router conditions, terminal outcomes, and input contract — over the workflows MCP write ops.
/workflows:build-skillsInspect every live workflow and reconcile project-local agent operations for interactive, automatic, routed, and human-intervention work.
/workflows:connectUse when connecting a repository and machine to a Workflows board.
/workflows:discussTalk to one workflow item — load its documents, Q&A history, and place in its pipeline, discuss it with the user, and carry any amendment back through the safest path.
/workflows:run-itemDrive a workflow item through its autonomous transitions from THIS interactive session — the next_step → execute → submit_step loop — instead of the headless engine.
/workflows:unblockInspect the configured project's human-intervention queue, handle selected items in order, or guide one waiting item at a time without taking a run claim.
/workflows:writeApply the system's writing-craft guidance — lead with the conclusion, visualize, write for the cold-start reader, be concise — to whatever you're about to write.
workflowsruns: bun run ${CLAUDE_PLUGIN_ROOT}/mcp.js
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.