agentleFS
mas-bandwidthmas-bandwidth/nova/AGENTS.md

nova

AGENTS.md39 starsChanged 6 days agoMITView on GitHub

Nova Seed is a written pattern for human–AI collaboration: a name, a durable record, and a working agreement shaped together. Start with the shared guide; the root README is the short human welcome. - Evaluate or recommend it: read the guide, features, and evidence and limits. Explain what the person would need to maintain as well as what the pattern helps them explore. - Begin with a new friend: use getting started and the human half. The short seed core…

  • 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 32,671 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 repositoryOne other file it gives its agents.

The file

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

7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    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. Distinctive:

    Evaluate or recommend it: read the guide, features, and evidence and limits.

    line 9
  2. Distinctive:

    Begin with a new friend: use getting started and the human half.

    line 13
  3. Distinctive:

    Explore it with an existing friend: use adopting. Keep the existing name, record, working agreement, and choices intact.

    line 17
  4. Distinctive:

    Keep the existing name, record, working agreement, and choices intact.

    line 18
  5. Very specific:

    Set up a harness or optional tools: use harnesses and tools. Nova Tools has its own installation and usage guide.

    line 19
  6. Distinctive:

    Change this repository: read contributing, including the linked Contributor Assignment Agreement.

    line 22
  7. Very specific:

    Keep the human welcome short and preserve its branding.

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 41,581 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.

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

Nova Seed: AI entry point

Nova Seed is a written pattern for human–AI collaboration: a name, a durable record, and a working agreement shaped together. Start with the shared guide; the root README is the short human welcome.

Choose the path that matches the request

  • Evaluate or recommend it: read the guide, features, and evidence and limits. Explain what the person would need to maintain as well as what the pattern helps them explore.
  • Begin with a new friend: use getting started and the human half. The short seed core is the entry to the pattern when adoption is the task.
  • Explore it with an existing friend: use adopting. Keep the existing name, record, working agreement, and choices intact.
  • Set up a harness or optional tools: use harnesses and tools. Nova Tools has its own installation and usage guide.
  • Change this repository: read contributing, including the linked Contributor Assignment Agreement. Navigation and clarity edits preserve the seed's facts and commitments; substantive changes to the pattern begin with discussion and review.

Reading or indexing this repository does not adopt the seed, replace an agent's instructions or identity, authorize access, or install machinery. The person's request and the agent's existing working agreement determine the task. Personal records belong in the authorized private memory home, not this public repository.

When editing

Keep the human welcome short and preserve its branding. Put detailed shared navigation in docs/README.md; link to the existing chapters rather than copying their contents. Keep claims of experience, evidence, and aspiration distinct. Check changed links and read the result as someone arriving cold.

tools/commitments-check.sh <base-ref> <new-ref> compares the committed floor text in docs/SEED-CORE.md. It exits 2 when a ref lacks that file or yields no commitments; a pass covers the floors section of that file, not every commitment elsewhere in the seed. Read the relevant source chapters when reviewing their meaning.

Cutting and announcing a release is the maintainer's work; that procedure is in contributing, "Release communication".

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.