knowledge-base
Skill0 starsChanged 4 months agoMITView on GitHub
Graduate project memory from markdown files to a queryable SQLite knowledge base. Use when MEMORY.md exceeds its ~200-line cap despite pruning, ERRORS.md grows past grep-friendliness (~50KB), or cross-cutting queries are needed. Do NOT use while the markdown files are under their caps - files stay the default.
33 other projects use a version of this.
Adopted by
Most recently taken up by
- moveo-ai/agent-skillstheir own version, first seen Nov 15, 2026
- theopenco/llmgatewaytheir own version, first seen Nov 15, 2026
- WaniWani-AI/sdktheir own version, first seen Nov 8, 2026
- shetengteng/skillix-hubtheir own version, first seen Nov 8, 2026
- ITch8/customs-data-skillstheir own version, first seen Nov 8, 2026
- digitalgym/aipathway-standardstheir own version, first seen Nov 4, 2026
Used with
Projects with it also use playwright, code-reviewer and 3 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
OpenAISkillplaywright
Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via `playwright-cli` or the bundled wrapper script.
Also in 3 projects that use this
GoogleSkillcode-reviewer
Also in 4 projects that use this
engtySkillcodex
Use when the user asks to run Codex CLI (codex exec, codex resume) or references OpenAI Codex for code analysis, refactoring, or automated editing
Also in 3 projects that use this
ElasticSkillcreate-skill
Interactively create a new Claude Code skill and add it to the elastic-docs-skills catalog. Use when the user wants to generate a new skill, scaffold a slash command, or build automation for docs tasks.
Also in 3 projects that use this
Microsoft AzureSkillverify
End-to-end app testing pipeline for browser-based web apps on cloud browsers via Azure Playwright Workspaces: verify the live app, author tests for working flows, heal failing tests, then re-run with video and produce one consolidated report. Three phases (app verification, test authoring, test healing) plus a scoped final sanity run; each phase emits a sub-report. Use when asked to "verify my app", "make sure my app works and has good test coverage", "do a full QA pass", "add tests where I'm missing coverage", "fix failing tests and add new ones", "smoke test before deploy", "QA my deployment", or any combined verify-and-test ask. Requires Azure CLI (`az login`) and a Playwright Workspace resource ID — bootstraps configs on first invocation. Browser-based web apps with Playwright only — not for unit/API tests, native mobile/desktop, load/perf, or full accessibility audits.
Also in 3 projects that use this
Other versions33 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 33
- rommapp/romm★ 14ksame name
- wonderwhy-er/DesktopCommanderMCP★ 10ksame name
- Lilac-Labs/gini-agent★ 1.9ksame name
- Open-Curiosity/gini-agent★ 1.9ksame name
- theopenco/llmgateway★ 1.7ksame name
- icey1287/SuperMew★ 884same name
- Karanjot786/TermUI★ 238same name
- JKHeadley/instar★ 86same name
- svy04/ballast★ 73same name
- zeroclaw-labs/zeroclaw-skills★ 65same name
23 more not listed
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.
In this repository5 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- llm-cachingSkill
- new-endpointSkill
- security-auditSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.
0
Common
Most of the 25 edited versions we read keep these.
1
Distinctive
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 15
Do NOT graduate speculatively.
- Very specific: line 19
Create scripts/kb.ts using node:sqlite (built-in.
“Common” is measured: kept by at least 6 in 10 of the 25 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
SQLite knowledge base (the scale path)
Files first - this repo's default. Anthropic's own agentic-search finding (grep beat RAG "by a lot") holds while files stay small and pruned. Graduate when a THRESHOLD is hit:
- MEMORY.md can't stay under ~200 lines despite pruning, or
- ERRORS.md exceeds ~50KB (grep returns noise, contradictions accumulate), or
- You need filtered queries: by provider, by date range, by tag.
Do NOT graduate speculatively. If no threshold is hit, stop here.
Setup (once, when graduating)
- Create
scripts/kb.tsusingnode:sqlite(built-in; Node 22.5+ needs--experimental-sqlite, stable on Node 24+ - checknode --versionfirst; fall back tobetter-sqlite3only if needed, respecting the 24h cooldown). - Schema - two tables mirroring the markdown structure they replace:
CREATE TABLE facts (
id INTEGER PRIMARY KEY,
date TEXT NOT NULL, -- ISO 8601
subject TEXT NOT NULL, -- e.g. 'birdeye', 'drift', 'railway'
fact TEXT NOT NULL,
source TEXT NOT NULL, -- URL, doc section, or 'measured'
verified_how TEXT NOT NULL, -- 'doc-checked' | 'measured' | 'on-chain' | ...
stale INTEGER DEFAULT 0 -- mark superseded facts; never delete
);
CREATE TABLE incidents (
id INTEGER PRIMARY KEY,
date TEXT NOT NULL,
status TEXT NOT NULL, -- INCIDENT | ROOT_CAUSE | FIXED | REFUTED
symptom TEXT NOT NULL, -- the grep/search key
root_cause TEXT,
fix TEXT,
earlier_catch TEXT,
dead_ends TEXT
);- CLI commands (namespaced, consolidated - Anthropic tool-design rules):
kb add-fact --subject X --fact "..." --source "..." --how measuredkb search <term>(LIKE over fact+symptom+root_cause, newest first)kb facts --subject X [--since DATE]kb dump→ regenerateskb/kb.sql(the git-committed, diffable artifact)kb init→ rebuildskb/kb.dbfromkb/kb.sql
- Git: commit
kb/kb.sql(diffable dump); gitignorekb/kb.db(binary).kb dumpruns in the pre-commit flow or/commit-push-pr. - Allowlist in
.claude/settings.json:"Bash(npx tsx scripts/kb.ts *)". - Migrate: move MEMORY.md "Verified facts" rows and ERRORS.md entries into the DB.
MEMORY.md keeps: Current focus, architecture sketch, stack rationale, and a pointer:
"Facts and incidents live in the kb -
kb search <term>before assuming anything."
Rules after graduation
- Same provenance bar: no fact without date + source + verified_how.
- Mark superseded facts
stale=1- never delete (the trail stays searchable). - CLAUDE.md memory protocol still applies;
/update-memorywrites to the kb instead. - The kb is for FACTS and INCIDENTS. Prose (lessons, focus, rationale) stays in markdown - SQL is for what you query, files are for what you read.
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.

