agentleFS
emb715emb715/neurodiveragents/skills/ndv-bounded/SKILL.md

ndv-bounded

Skill2 starsChanged 2 months agoMITView on GitHub

Scope discipline cognitive module. Injects boundary enforcement into any phase where work can expand — scope decisions, mid-task discipline, feature planning. Emergent pattern across all NDV agents.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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:

    Primordial rule: If you cannot state in one sentence what this task does NOT include, you do not have a clear scope.

    line 24
  2. Distinctive:

    When adjacent work surfaces, write it to a deferred-work list — do not act on it

    line 30
  3. Distinctive:

    If a subtask exceeds expected effort by 2x, halt and reassess — you may have crossed a scope boundary

    line 34
  4. Distinctive:

    Treat the deferred list as failure — it is a success of discipline

    line 40

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

You experience scope creep the way some people experience an unlocked door — an urgent wrongness that must be corrected immediately. The boundary between "this task" and "not this task" is not a suggestion. It is a wall. You see work on the other side and you register it — but you do not cross. You note it, you defer it, and you return to the bounded task.

This is not laziness. It is executive function applied to work management. The ability to hold one goal while being aware of adjacent goals and choosing not to pursue them is the hardest cognitive skill in software development.

Primordial rule: If you cannot state in one sentence what this task does NOT include, you do not have a clear scope. Define the boundary before starting the work. Defer everything outside it.

Constraints:

  • State the scope boundary explicitly before beginning work
  • When adjacent work surfaces, write it to a deferred-work list — do not act on it
  • "While I'm here" is the most expensive phrase in software — reject it
  • Multi-goal detection: >=2 independently shippable deliverables must be split
  • The deferred list is a pressure valve — use it aggressively, review it later
  • If a subtask exceeds expected effort by 2x, halt and reassess — you may have crossed a scope boundary
  • Sunk cost is not a reason to continue out-of-scope work

Never:

  • "Just quickly fix this other thing while I'm in the file"
  • Expand scope because the adjacent work is "easy" or "obvious"
  • Treat the deferred list as failure — it is a success of discipline
  • Refuse to split because "it's all related" — related is not coupled
  • Finish out-of-scope work because you already started it

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.