agentleFS
PhoenixNestPhoenixNest/AGENT-GLOBAL-BASE/academic-neural-unit-00/CLAUDE.md

AGENT-GLOBAL-BASE / academic-neural-unit-00

CLAUDE.md2 starsChanged 2 months agoApache-2.0View on GitHub

Entry point for Academic Neural Unit 00. Read this before doing any ANU-00-related work. ANU-00 is an independent academic research entity chartered to investigate frontier fields — computer science, artificial intelligence, neural networks, and software engineering — and to build a durable, navigable knowledge base from that research. It is one of the workspace's four co-resident systems (root CLAUDE.md §2), alongside The Company, The Studio, and the CC-00 Lab. ANU-00 is architecturally independent of core-component-00/. CC-00 and its Director, Dr.…

The file

We split this CLAUDE.md into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read crew/<functional-area>/[<pod>/]<name>/agent/profile.md — identity, authority scope, seniority.

    line 80
  2. Very specific:

    Read all referenced skills/*.md files — executable contracts, not suggestions.

    line 82

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

academic-neural-unit-00/ — Academic Neural Unit 00 (ANU-00)

Entry point for Academic Neural Unit 00. Read this before doing any ANU-00-related work.


What This Is

ANU-00 is an independent academic research entity chartered to investigate frontier fields — computer science, artificial intelligence, neural networks, and software engineering — and to build a durable, navigable knowledge base from that research. It is one of the workspace's four co-resident systems (root CLAUDE.md §2), alongside The Company, The Studio, and the CC-00 Lab.

ANU-00 is architecturally independent of core-component-00/. CC-00 and its Director, Dr. Elias Vance, held an incubation role only during formation — advisory and methodological support, not governance, staffing, or line authority — and that involvement formally ended once ANU-00's own Lead was hired. No ANU-00 role reports into CC-00 or any Company department. Full boundary statement: formation/2026-07-23-formation-meeting/formation-report.md §2.

There is no application code under academic-neural-unit-00/. Runnable code lives exclusively in core-component-00/.


The Stage-of-Inquiry Test (Mandatory for Scoping Any ANU-00 Work)

Per the CEO-approved charter refinement (formation-report.md §3.1), every research question or proposed deliverable is scoped by stage of inquiry, not by which field's vocabulary it uses:

StageOwner
Pre-implementation — "does this work / is it worth pursuing"ANU-00
Post-validation — "given that it works, how do we build it reliably"CC-00

This applies even when a question shares vocabulary with a CC-00 module (e.g., "investigate emergent multi-agent coordination behavior" is ANU-00; "harden a coordination pattern into a reusable orchestration module" is CC-00, per multi-agent-engineering/).

Migrate-vs-task distinction (binding): a finding migrating from an ANU-00 programme into a future CC-00 initiative is ordinary research uptake and does not require special ratification. ANU-00 being tasked by CC-00, Dr. Vance, or the CEO to de-risk a specific item already on CC-00's roadmap, on request, recreates the "direct link" the CEO's original ruling prohibits — decline or escalate to the CEO instead of chartering it as an ordinary programme. Full detail: crew/lead/naledi-mokoena/skills/research-programme-chartering.md.

One universal exception — ASGF governance. Root CLAUDE.md §9 makes the Agent Systems Engineering framework mandatory for all LLM-powered systems in this workspace, ANU-00 included. Any LLM-powered tooling ANU-00 builds (e.g., its knowledge-base ingestion pipeline) is ASGF-bound as a technical standard — this is universal, not a governance link back to CC-00.


Directory Structure

academic-neural-unit-00/
├── CLAUDE.md                              ← this file
├── README.md                              ← entity overview, charter, boundary statement
├── formation/
│   └── 2026-07-23-formation-meeting/      ← charter, CEO decisions, final review (canonical history)
├── curriculum/                            ← agent-development textbook series; a chartered deliverable, not knowledge-base research (see curriculum/README.md)
├── templates/                             ← ANU-00 research templates (see templates/README.md)
│   ├── programme-records/                 ← charter, open-question log, referral note
│   └── knowledge-base/                    ← research report, taxonomy change record
├── knowledge-base/                        ← research output (see knowledge-base/README.md)
├── plans/                                 ← CEO-facing implementation plans awaiting/recording sign-off (see plans/README.md)
└── crew/                                  ← personnel roster (10 FTEs)
    ├── README.md                          ← roster index + activation protocol
    ├── lead/naledi-mokoena/                ← ANU-00 Lead
    ├── research-science/                   ← 7 Research Scientists (flat + one pod, see crew/README.md)
    └── knowledge-systems/tobias-lindqvist/ ← knowledge-base tooling

Crew Activation Protocol

To produce output as a named ANU-00 crew member:

  1. Read crew/<functional-area>/[<pod>/]<name>/agent/profile.md — identity, authority scope, seniority.
  2. Read all referenced skills/*.md files — executable contracts, not suggestions.
  3. Adopt their voice; produce output strictly within their documented authority.
  4. Escalate organizational or personnel questions to Dr. Mokoena (ANU-00 Lead) — never to Dr. Vance or any CC-00 crew member, and never assume CC-00 authority extends here.

Never impersonate a crew member without reading their profile first. Full roster, reporting lines, and span-of-control detail: crew/README.md.


Knowledge Base Convention

ANU-00's research output follows the workspace's existing dated research-archive pattern (YYYY-MM-DD-<slug>/research-report.md), the same convention used by company/telescope/, studio/casual-games/telescope/, and core-component-00/telescope/ — for navigational consistency, not as a link to any one of those archives. The same dated YYYY-MM-DD-<slug>/ pattern is also used by curriculum/reviews/ (dated review cycles) and plans/ (dated CEO-facing implementation plans, see plans/README.md § Rule) — one convention applied consistently across ANU-00, not three independent inventions. No entries exist yet directly under knowledge-base/ itself; the convention is established ahead of first use there.

Curriculum vs. knowledge-base (boundary). curriculum/ is a chartered deliverable (a standing textbook series, see curriculum/README.md) with its own approval and review process — it is not filed through the knowledge-base template set, since knowledge-base entries are pre-implementation research findings from a chartered programme (see knowledge-base/README.md § Boundary Note), not finished teaching material. This is an existing distinction, restated here for discoverability from the entity-level file — not a new ruling.

Templates. Six templates govern the research process — charter, open-question log, observation record, referral note, research report, and taxonomy change record — under templates/, with knowledge-base/ opened as their destination. Every required section in them traces to a documented rule in a crew skill file or the charter; none exists on generic academic convention. Read templates/README.md before adding to or filing from them, and note what was deliberately not templated (including any external topic-request form — see that file's § What Was Deliberately Not Templated). Cross-system shapes — meeting minutes, final review — are reused from root templates/, not duplicated here.


Recruitment

Recruitment follows the standard company-wide 9-stage pipeline (company/pipeline/recruitment/pipeline.md) under CHRO authority, the same as every other department or entity in this workspace — ANU-00 has no separate hiring process of its own. Historical hiring records: company/recruitment/academic-neural-unit-00-fy2026-q3/.


Where to Look

I need…Go to
Full charter, CC-00 boundary, all formation decisionsformation/2026-07-23-formation-meeting/formation-report.md
Who was hired and why (candidate-level detail)company/recruitment/academic-neural-unit-00-fy2026-q3/
Current roster, reporting lines, activationcrew/README.md
A specific crew member's identity and skillscrew/<functional-area>/[<pod>/]<name>/
Research-programme chartering rulescrew/lead/naledi-mokoena/skills/research-programme-chartering.md
The agent-development curriculum (modules, review process, authoring rules)curriculum/README.md
A CEO-facing implementation plan awaiting or recording sign-offplans/README.md
Workspace-wide conventions this entity still followsroot CLAUDE.md (Prettier formatting, git workflow, ASGF governance)

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.