Claude Skill

brief-first-execution

Create a single working brief before substantial implementation, audit, or migration work. Use when tasks are non-trivial, risks are real, or the session is likely to drift without one source of truth for scope, non-goals, verification, and done criteria.

LLM Mart · 0 points · 3 views 0 listing impressions 0 install-command copies
Download markoblogo-abvx-agent-skills-skills_brief-first-execution-37a7146.zip · 2 KB

Install

skills CLI npx skills add https://github.com/markoblogo/abvx-agent-skills/tree/main/skills/brief-first-execution
Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install markoblogo-abvx-agent-skills@llmmart
Git git clone https://github.com/markoblogo/abvx-agent-skills.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole markoblogo/abvx-agent-skills collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

Brief First Execution

Start non-trivial work with one live task brief.

Goal

Reduce drift by creating a compact source of truth for:

  • scope;
  • non-goals;
  • affected areas;
  • risks;
  • verification;
  • done criteria.

When To Use

Use for:

  • multi-file implementation;
  • behavior changes;
  • migrations, audits, or refactors;
  • long runs where context could drift;
  • work with real operational or delivery risk.

Skip for:

  • tiny one-file edits with obvious blast radius;
  • trivial questions that do not create lasting change.

Workflow

  1. Name the task
    • write a short title that matches the real delivery target.
  2. Define scope
    • describe what is in;
    • keep it concrete and delivery-shaped.
  3. Define non-goals
    • state what this task will not do;
    • use this to stop opportunistic scope creep.
  4. List affected areas
    • identify modules, docs, data flows, or systems likely to change.
  5. List risks
    • focus on failure modes that matter for this task.
  6. Define verification early
    • list the narrowest useful checks first;
    • add broader checks only where the blast radius warrants them.
  7. Define done criteria
    • describe what must be true before closing the task.
  8. Keep the brief live
    • update the brief when scope changes materially;
    • do not leave the brief frozen while the work mutates underneath it.

Rules

  • One task, one brief.
  • Scope and non-goals should both be explicit.
  • Verification belongs in the brief before heavy implementation starts.
  • If the task changes enough, revise the brief rather than pretending it did not.

Final Report

Return the brief itself and call out any unresolved scope or verification questions.

Files (abvx-agent-skills)
  • agents
    • openai.yaml 218 B
      Loading…
  • SKILL.md 2.1 KB
    ---
    name: brief-first-execution
    description: Create a single working brief before substantial implementation, audit, or migration work. Use when tasks are non-trivial, risks are real, or the session is likely to drift without one source of truth for scope, non-goals, verification, and done criteria.
    license: MIT
    metadata:
      abvx_status: experimental
      abvx_origin: original
    ---
    
    # Brief First Execution
    
    Start non-trivial work with one live task brief.
    
    ## Goal
    
    Reduce drift by creating a compact source of truth for:
    
    - scope;
    - non-goals;
    - affected areas;
    - risks;
    - verification;
    - done criteria.
    
    ## When To Use
    
    Use for:
    
    - multi-file implementation;
    - behavior changes;
    - migrations, audits, or refactors;
    - long runs where context could drift;
    - work with real operational or delivery risk.
    
    Skip for:
    
    - tiny one-file edits with obvious blast radius;
    - trivial questions that do not create lasting change.
    
    ## Workflow
    
    1. **Name the task**
       - write a short title that matches the real delivery target.
    2. **Define scope**
       - describe what is in;
       - keep it concrete and delivery-shaped.
    3. **Define non-goals**
       - state what this task will not do;
       - use this to stop opportunistic scope creep.
    4. **List affected areas**
       - identify modules, docs, data flows, or systems likely to change.
    5. **List risks**
       - focus on failure modes that matter for this task.
    6. **Define verification early**
       - list the narrowest useful checks first;
       - add broader checks only where the blast radius warrants them.
    7. **Define done criteria**
       - describe what must be true before closing the task.
    8. **Keep the brief live**
       - update the brief when scope changes materially;
       - do not leave the brief frozen while the work mutates underneath it.
    
    ## Rules
    
    - One task, one brief.
    - Scope and non-goals should both be explicit.
    - Verification belongs in the brief before heavy implementation starts.
    - If the task changes enough, revise the brief rather than pretending it did not.
    
    ## Final Report
    
    Return the brief itself and call out any unresolved scope or verification questions.
    
  • SKILL_CARD.md 1.3 KB
    Loading…

Comments (0)

Sign in to join the conversation.

No comments yet.

Reviews (0)

No reviews yet.