Workflow
Plan & Build
Research and plan complex work, then build it with an evidence-gated team.
Research → plan → spec → build, end to end.
kr 2 700View plugin
Make Claude Code reason like your most disciplined engineer.
Four reasoning disciplines that fire on their own at the right moments — Claude explores before committing, reasons in the open, proves work is done before claiming it, and finds the root cause before touching a fix. A session-start primer makes Claude reach for them, so you get the rigour without remembering to ask.
- Disciplines auto-fire on triggers
- No API keys, no Python
- Nothing leaves your machine
kr 850one-time · lifetime updates · install →
What runs when you install the plugin.
Five mechanisms, no loop to run. Install it and the disciplines arm themselves — each fires on its own trigger; only setup is ever manual.
- 01Idea Shaping turns a rough idea into a validated design — one question at a time, alternatives explored before you commit. The design is captured to your workspace.
- 02Structured Reasoning makes the thinking explicit — analysis → reasoning → decision, in the open, on every non-trivial call.
- 03Completion Proof blocks every "done / fixed / passing" claim that isn't backed by fresh evidence gathered in the same message.
- 04Root-Cause Analysis finds the real cause first — disciplined investigation instead of guess-and-check fixes.
- 05A session-start primer arms all four and their triggers the moment a session opens, so Claude reaches for them on its own.
Every artefact you're getting.
A plugin is a folder. Here's everything in it — four skills, one command, one hook. Nothing hidden, nothing abstract.
4 Skills
Atomic disciplines Claude Code loads on cue, isolated context.
idea-shapingTurns a rough idea into a validated design — one question at a time, alternatives first.
structured-reasoningMakes non-trivial thinking explicit: analysis → reasoning → decision.
completion-proofBlocks any "done / fixed / passing" claim not backed by fresh evidence.
root-cause-analysisFinds the real cause before any fix — investigation over guess-and-check.
The whole plugin sits at ~/.claude/plugins/core-methodology/ — every discipline is a file you can read, edit, and version.
Will it work with your stack?
Reads from
Your Claude Code session
Writes to
Plugin memoryworkspace/core-methodology/
Requirements: Claude Code (Anthropic's free CLI). No API keys, no Python, no uv — the lightest install in the catalogue.
No integrations to wire. It reads nothing external and runs entirely in-session — the only files it writes are your own design captures.
Four steps. From buy to first run.
Buy on the right. On purchase, we invite your GitHub account to the private plugin repo you bought — the Tenkr marketplace itself is public — and email you the two commands below.
1Install Claude Code (Anthropic's free CLI). Already have it? Skip.
2Buy. We invite your GitHub account to the private plugin repo you paid for. The Tenkr marketplace itself is public.
3In Claude Code, run
/plugin marketplace add TenkrAS/tenkr-plugin-marketplace to register the public catalogue. Exact command in your purchase email.4Run
/plugin install <plugin>@tenkr-plugin-marketplace — Claude Code clones it from your private repo and you're live. Type the plugin's command to start.What people ask before buying.
/ 01Do I have to run a command every time?
No. The four skills auto-fire on their own triggers; only
setup is manual, and it's safe to re-run./ 02Will it change my files?
No.
setup only records canonical decisions to plugin memory. The one thing that writes is Idea Shaping, which saves each validated design to workspace/core-methodology/./ 03Does it send my code anywhere?
No. It runs locally with your own Claude key and makes no external calls. Tenkr never sees what it touched.
/ 04How is this different from just prompting well?
The disciplines fire automatically at the right moments — a session-start primer makes Claude reach for them, so you get the rigour without remembering to ask for it.