agentleFS
gouthamk16gouthamk16/AttogradDB/docs/agents.md

AttogradDB

AGENTS.md4 starsChanged 3 months agoMITView on GitHub

Install AttogradDB project memory in Cursor and Claude Code from the plugin marketplace

  • 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 repository2 other files it gives its agents.

The file

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

1 rule. About 0 are common, 1 is distinctive, and 0 are 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Prefer the Python API instead of an agent?

    line 63

≈ “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

Claude Code and Cursor

Cursor and Claude Code load AttogradDB as a plugin — one click or one command, no config files to edit. The plugin runs a per-project decision-memory server and exposes two tools:

  • recall_decisions — every active decision for this project, in order.
  • remember_decision — record a durable choice, superseding an older one when it replaces it.

Each project keeps its own .attograd-memory.db. The plugin passes the active workspace to the server, so you never type a project path.

Using another tool (Codex, Gemini CLI, OpenCode, or anything else)? See Every other tool.

Cursor

Install from the Cursor plugin directory: cursor.directory/plugins/attograd-memory.

  1. Open the listing and click Install, or in Cursor open Settings → Plugins, search for attograd-memory, and install it.
  2. Choose the global (user) scope so it applies to every project.
  3. Confirm attograd-memory is enabled under Settings → MCP, then start a new Agent chat.

Cursor passes ${workspaceFolder} to the server, so memory is scoped to whichever project you have open.

Claude Code

The plugin installs from a marketplace inside a Claude Code session:

/plugin marketplace add gouthamk16/AttogradDB
/plugin install attograd-memory@attograd-plugins

Pick the global/user scope when prompted. Run /mcp to confirm attograd-memory connected, then start a new session. Claude Code passes the active project directory to the server.

Once the plugin is approved in Anthropic's community directory it also installs as attograd-memory@claude-community (/plugin marketplace add anthropics/claude-plugins-community). Both give you the same tools and the same per-project database.

Using it

The bundled skill tells the agent to call recall_decisions before it plans and to record durable choices with remember_decision. You can also ask directly:

  • "Recall the project decisions before you start."
  • "Remember that we use SQLite, not Redis, for durability — supersede the old Redis decision."

remember_decision takes claim, rationale, and optional scope, evidence, and supersedes (the id of the decision it replaces). A missing, cross-project, or already-superseded supersedes target is rejected without a partial write.

Prefer the Python API instead of an agent? See Python usage. How listings get submitted is in Marketplace listings.

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.