kb-init
Skill15 starsChanged 24 days agoMITView on GitHub
Start a Pointframe session - print the knowledge base core, the workflow rules for your role, environment facts, and optionally the sections and lessons for a file, feature, or topic. Use when the user types /kb-init; at the start of every session, before any task; when your role or your target changes.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- smoketurner/claude-okf-kbtheir own version, first seen Nov 8, 2026
- jsobolewski1/fonduetheir own version, first seen Oct 13, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- jsobolewski1/fondue★ 6same name
- smoketurner/claude-okf-kb★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 26
Read the output. The rules and the core apply to everything you do in this session.
- Very specific: line 27
Read each workflow section it names for your role, and the sections of the target it lists, at the lines it gives.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/kb-init: start every session here
/kb-init [coordinator|executor|reviewer|helper] [target] prints what a session needs, in one command:
pwsh scripts/kb.ps1 init [-Role coordinator|executor|reviewer|helper] [target]The role is the first argument when it is one of the four words, otherwise executor. The coordinating session that plans, briefs, reviews, and merges uses coordinator; an agent that only checks a result uses reviewer; a read-only Haiku helper uses helper. A subagent cannot type slash commands, so a coordinator starts every helper by telling it in the brief to run pwsh scripts/kb.ps1 init -Role helper, or pastes that output into the brief. The target is optional: a file path, an F-NN feature ID, an area, or a topic such as dpi.
It prints, in order:
- The whole core,
docs/knowledge-base/knowledge-base.md: architecture rules, the Feature and Decision indexes, and links to composition, invariants, and how-tos. - The workflow rules for your role, verbatim, with their
W-NNIDs, and the names of the extra workflow sections your role should read. - Environment facts from read-only probes (a running
Pointframe.Mcp.exe, the desktop lock, hidden trial installs, the worktree count). They are facts, not instructions;unknownmeans a probe failed. - Coordinator only: the active records of
plan/status.json(objective, stage, next, blocker, due), labelled local operational context and not policy. - With a target: the
kb readoutput, then the full text of the sections it names and its top lessons. A feature or topic that matches more than three sections prints a menu (heading, file:line, size); open the ones you need.
Then:
- Read the output. The rules and the core apply to everything you do in this session.
- Read each workflow section it names for your role, and the sections of the target it lists, at the lines it gives.
/kb-read <file, area, F-NN, or topic>finds more later. - The last line shows bytes and estimated tokens per part; an overage line means a part passed its budget.
Claude Code can run this at the start of every session with an optional SessionStart hook; see "Maintenance commands" in docs/knowledge-base/maintenance.md.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

