agentleFS
gouthamk16gouthamk16/AttogradDB/skills/attograd-memory/SKILL.md

attograd-memory

Skill4 starsChanged 3 months agoMITView on GitHub

Recall and record durable project decisions with AttogradDB. Use when starting work in a repository, planning or editing code, or deciding whether a project constraint should persist.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 5 are distinctive, and 3 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.

  • 3

    Very specific

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

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

    Use the attograd-memory MCP tools for decisions that should survive the current task.

    line 8
  2. Distinctive:

    Before planning or editing:

    line 12
  3. Distinctive:

    Treat every returned active decision as a project constraint.

    line 15
  4. Distinctive:

    Do not contradict an active decision without an explicit replacement.

    line 16
  5. Very specific:

    Call remember_decision when the user or team makes a durable choice about architecture, dependencies, APIs, workflow, security, or scope.

    line 20
  6. Very specific:

    When a new decision replaces an active one, pass the old decision's id as supersedes.

    line 28
  7. Distinctive:

    Do not create a contradictory active decision.

    line 29
  8. Distinctive:

    Do not store transient task details, guesses, or secrets.

    line 31

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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
SKILL.md

AttogradDB project memory

Use the attograd-memory MCP tools for decisions that should survive the current task.

Start of a task

Before planning or editing:

  1. Call recall_decisions.
  2. Treat every returned active decision as a project constraint.
  3. Do not contradict an active decision without an explicit replacement.

Record durable decisions

Call remember_decision when the user or team makes a durable choice about architecture, dependencies, APIs, workflow, security, or scope. Include:

  • claim: the decision in one clear sentence.
  • rationale: why it was chosen.
  • scope: the affected area when useful.
  • evidence: a file, benchmark, issue, or other supporting reference when available.

When a new decision replaces an active one, pass the old decision's id as supersedes. Do not create a contradictory active decision.

Do not store transient task details, guesses, or secrets.

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.