maz-works-knowledge-vault
AGENTS.md0 starsChanged 3 months agoMITView on GitHub
This is the single shared knowledge vault for Maz Works and Manazir Hussain. JobFilter is one project inside the vault. Do not describe the complete vault as a JobFilter vault and do not create another general-purpose vault. Before project work: At session end: Skills live at skills/<name>/SKILL.md. Multi-agent discovery uses the neutral package path: Run bash bin/setup-multi-agent.sh to install missing links. The repository includes Obsidian knowledge workflows, hybrid retrieval, methodology routing, per-file write locking, source ingestion, wiki health checks, session…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,439 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository18 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- .cursor/rules/maz-works-knowledge-vault.mdcCursor rule
- autoresearchSkill
- canvasSkill
- defuddleSkill
- obsidian-basesSkill
- obsidian-markdownSkill
- saveSkill
- thinkSkill
- wiki-cliSkill
- wiki-foldSkill
- wiki-ingestSkill
- wiki-lintSkill
- wiki-modeSkill
- wiki-querySkill
- wiki-retrieveSkill
- wikiSkill
The file
We split this AGENTS.md into the rules it gives an agent
9 rules. About 1 is common, 2 are distinctive, and 6 are very specific to this project.
≈1
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Very specific: line 7
YouTube URL from Maz → video intelligence: follow prompts/youtube-video-intelligence.md (full transcript, facts vs opinion vs inference, …
- Very specific: line 34
Read the relevant wiki/projects/{project}/ context.
- Very specific: line 35
Read wiki/hot.md only when cross-project context is useful.
- Very specific: line 36
Read the complete SKILL.md for any selected skill before using it.
- Distinctive: line 40
Update the relevant project status or handoff.
- Very specific: line 42
Commit and push the vault to fork main.
- Distinctive: line 49
Use Obsidian wikilinks for internal knowledge links.
- Common: line 51
Never commit credentials, private client data, confidential material, or private spiritual content.
- Very specific: line 63
Run bash bin/setup-multi-agent.sh to install missing links.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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
Maz Works Knowledge Vault — Agent Instructions
Maz Works sales: start here (every agent)
- Focus:
NOW.md(3 projects only) - Lead rules: Lead Quality v2 (problem → evidence → paid intervention; gates; Gold/Silver/Bench; Gold does not need a Ltd) + hand-off format:
prompts/maz-works-niche-needs.md. Offers/prices only frommazos-site/app/offers.ts. - YouTube URL from Maz → video intelligence: follow
prompts/youtube-video-intelligence.md(full transcript, facts vs opinion vs inference, connect to our projects, save a note inwiki/sources/video-intelligence/, commitknowledge: ingest YouTube research - <title>). Searchwiki/sources/video-intelligence/_index.mdbefore new research or a related strategic decision. - Prompt library + /mw-* commands:
prompts/maz-works-prompt-library.md - LinkedIn posts:
prompts/maz-works-linkedin-launch.md - Leads, call list, pitches: PRIVATE repo https://github.com/manazoid4/maz-works-leads (local:
leads/, git-ignored here). Never copy lead data into this public repo. - Give leads to the owner as plain-text blocks (Google Docs format in niche-needs). Keep replies short.
Canonical vault
This is the single shared knowledge vault for Maz Works and Manazir Hussain.
- Canonical path:
C:\Users\manaz\Desktop\Maz Works Knowledge Vault - Repository:
https://github.com/manazoid4/maz-works-knowledge-vault - Main dashboard:
HOME.md - Current context:
wiki/hot.md - Shared index:
wiki/index.md - Project memory:
wiki/projects/ - Session handoffs:
wiki/sessions/ - Local ingestion:
Local Knowledge/ - Preserved history:
Archive/
JobFilter is one project inside the vault. Do not describe the complete vault as a JobFilter vault and do not create another general-purpose vault.
Session protocol
Before project work:
- Pull
mainfrom theforkremote. - Read the relevant
wiki/projects/{project}/context. - Read
wiki/hot.mdonly when cross-project context is useful. - Read the complete
SKILL.mdfor any selected skill before using it.
At session end:
- Update the relevant project status or handoff.
- Write
wiki/sessions/YYYY-MM-DD-{project}-{agent}.md. - Commit and push the vault to
fork main.
Knowledge rules
.raw/contains source material and is immutable.wiki/log.mdis append-only, newest entries first.wiki/hot.mdis refreshed rather than endlessly appended.- Use Obsidian wikilinks for internal knowledge links.
- Preserve sources, confidence, dates, and truth boundaries.
- Never commit credentials, private client data, confidential material, or private spiritual content.
Skills
Skills live at skills/<name>/SKILL.md. Multi-agent discovery uses the neutral package path:
~/.codex/skills/maz-works-knowledge-vault
~/.opencode/skills/maz-works-knowledge-vault
~/.gemini/skills/maz-works-knowledge-vaultRun bash bin/setup-multi-agent.sh to install missing links.
Maintained toolset
The repository includes Obsidian knowledge workflows, hybrid retrieval, methodology routing, per-file write locking, source ingestion, wiki health checks, session saving, research, and Canvas support. Run make test after changing scripts or skills.
The original MIT-licensed foundation is credited in ATTRIBUTION.md. Maz Works is the active identity and maintenance boundary.
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.

