agentleFS
vosslabvosslab/vosslab-skills/skills/planning/modest-plan-drafter/SKILL.md

modest-plan-drafter

Skill5 starsChanged 22 days agoMITView on GitHub

Draft concise plans for bounded coding, teaching, research, writing, or organizational work. Use when a short approach and observable completion checks suffice; use blueprint-plan-drafter for complex technical coordination.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use Modest when the outcome can be described, executed, and verified with a short approach.

    line 14
  2. Very specific:

    Use blueprint-plan-drafter for coding work involving cross-module coordination, shared contracts, migrations, significant compatibility …

    line 15
  3. Distinctive:

    Keep observations, assumptions, and decisions distinct.

    line 26
  4. Distinctive:

    Use this default shape, adapting it to the task:

    line 34

≈ “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

Modest plan drafter

Purpose and routing

Turn a request into a concise plan with evidence, scope, decisions, and observable outcomes. Produce the planning artifact before execution. Aim for roughly one page as a soft target; necessary decisions determine the detail.

Use Modest when the outcome can be described, executed, and verified with a short approach. Use blueprint-plan-drafter for coding work involving cross-module coordination, shared contracts, migrations, significant compatibility risk, or coordinated multi-agent implementation and review. Honor explicit skill choices. If a requested Modest plan leaves important technical decisions unresolved, explain the need for Blueprint or expand the relevant detail.

Establish the evidence

  • Inspect source material relevant to the outcome. For repository work, read operative repository rules or reuse a valid same-session receipt. Load specialized guidance only when applicable.
  • Establish the intended audience, current state, scope, and material constraints. Resolve discoverable facts from evidence; clarify choices that materially change the outcome.
  • Keep observations, assumptions, and decisions distinct. Use a bounded investigation with a decision rule when evidence is needed to select an approach.

For broad code investigation, read GRAPHIFY_GUIDE.md when a Graphify map or repository wrapper is available. Otherwise use targeted searches and source inspection.

Draft the plan

Use this default shape, adapting it to the task:

  1. Goal and context: intended outcome, relevant evidence, audience, and scope.
  2. Approach: a short sequence of meaningful actions, necessary decisions, and dependencies.
  3. Completion checks: observable evidence that the requested outcome is achieved.

For file-based work, add File scope after Approach: expected files or directories and the purpose of each change. Include relevant tests, documentation, and generated outputs with their canonical sources. Treat this as the expected edit boundary, refined as evidence emerges.

Include assumptions, constraints, tradeoffs, and material risks where useful. Name exact components, interfaces, or validation commands when they affect execution. Leave detailed task assignments and dispatch to execution. Keep the artifact proportional to the decisions the work needs.

Quality and verification

  • Ground requirements in product needs, meaningful contracts, measured constraints, or demonstrated failures. Use thresholds, byte or pixel equivalence, and exhaustive matrices only when those needs justify them. Improvement work should permit intended changes.
  • Apply KISS aggressively: mechanisms, abstractions, policies, state, configuration, and tests must solve demonstrated needs. Prefer sensible fixed behavior or a simpler shared design before adding options, parameters, modes, overrides, or extension points. Distinguish genuinely different tasks from internal implementation choices; every extra choice adds maintenance and combinations.
  • Give each new blocking gate a grounded reason and a concrete decision, correction, or recovery action on failure. Drop gates whose failure would identify no real problem or change no decision.
  • For software work, read CODE_VERIFICATION.md before drafting design and completion checks. It covers robustness, adaptability, repository policy, and the distinction between one-time evidence and justified permanent tests. For non-coding work, choose evidence appropriate to the deliverable.

Completion

Finish when the outcome and boundaries are clear, the approach carries necessary decisions and dependencies, file scope is useful where applicable, and completion checks are observable and proportionate. Resolve blocking choices or define an explicit evidence-led investigation that precedes dependent work. Return the plan in the requested format; save a file when requested or required by the active workflow. Execution begins as a separate task.

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.