Claude Skill

delivery-baseline-audit

Audit claimed delivery against the starting baseline and the current working tree. Use when a task has declared deliverables, when a long run might say 'done' before shipping real changes, or when final verification must check the actual artifact set rather than only the transcri

LLM Mart · 0 points · 2 views 0 listing impressions 0 install-command copies
Download markoblogo-abvx-agent-skills-skills_delivery-baseline-audit-37a7146.zip · 2 KB

Install

skills CLI npx skills add https://github.com/markoblogo/abvx-agent-skills/tree/main/skills/delivery-baseline-audit
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

Delivery Baseline Audit

Verify that the promised deliverables actually exist relative to where the work started.

Use When

  • the task names files, features, screens, commands, or other concrete deliverables;
  • the implementation ran through multiple phases or retries;
  • the working tree may contain staged, unstaged, or untracked work that matters;
  • "agent said done" is not enough evidence.

Core Rules

  • Compare against the starting baseline, not just the latest transcript claim.
  • Check the whole working tree, not only committed diffs.
  • Re-run final mandatory commands once at the end for cross-phase regressions.
  • Treat missing deliverables as a gap even if the narrative sounds complete.

Workflow

  1. Capture or identify the baseline reference before major work starts.
  2. List the declared deliverables and acceptance criteria.
  3. At the end, inspect the full working tree relative to that baseline:
    • committed changes;
    • staged changes;
    • unstaged changes;
    • deletions;
    • untracked deliverables.
  4. Re-run the final mandatory verification commands.
  5. Spot-check each deliverable against its declared acceptance criteria.
  6. Report gaps explicitly and fix them before calling the task complete.

Audit Output

Include:

  • baseline reference used;
  • deliverables checked;
  • commands re-run;
  • gaps found;
  • fix-needed or complete verdict.

Anti-Patterns

  • Do not audit only what was committed if the working tree still matters.
  • Do not rely on a prior mid-task pass as the final audit.
  • Do not mark complete when a declared deliverable is absent or unverifiable.

Final Report

State whether completion was confirmed against baseline, partially confirmed, or blocked by missing evidence.

Files (abvx-agent-skills)
  • agents
    • openai.yaml 222 B
      Loading…
  • SKILL.md 2.1 KB
    ---
    name: delivery-baseline-audit
    description: Audit claimed delivery against the starting baseline and the current working tree. Use when a task has declared deliverables, when a long run might say 'done' before shipping real changes, or when final verification must check the actual artifact set rather than only the transcript or commits.
    license: MIT
    metadata:
      abvx_status: experimental
      abvx_origin: adapted
    ---
    
    # Delivery Baseline Audit
    
    Verify that the promised deliverables actually exist relative to where the work started.
    
    ## Use When
    
    - the task names files, features, screens, commands, or other concrete deliverables;
    - the implementation ran through multiple phases or retries;
    - the working tree may contain staged, unstaged, or untracked work that matters;
    - "agent said done" is not enough evidence.
    
    ## Core Rules
    
    - Compare against the starting baseline, not just the latest transcript claim.
    - Check the whole working tree, not only committed diffs.
    - Re-run final mandatory commands once at the end for cross-phase regressions.
    - Treat missing deliverables as a gap even if the narrative sounds complete.
    
    ## Workflow
    
    1. Capture or identify the baseline reference before major work starts.
    2. List the declared deliverables and acceptance criteria.
    3. At the end, inspect the full working tree relative to that baseline:
       - committed changes;
       - staged changes;
       - unstaged changes;
       - deletions;
       - untracked deliverables.
    4. Re-run the final mandatory verification commands.
    5. Spot-check each deliverable against its declared acceptance criteria.
    6. Report gaps explicitly and fix them before calling the task complete.
    
    ## Audit Output
    
    Include:
    
    - baseline reference used;
    - deliverables checked;
    - commands re-run;
    - gaps found;
    - fix-needed or complete verdict.
    
    ## Anti-Patterns
    
    - Do not audit only what was committed if the working tree still matters.
    - Do not rely on a prior mid-task pass as the final audit.
    - Do not mark complete when a declared deliverable is absent or unverifiable.
    
    ## Final Report
    
    State whether completion was confirmed against baseline, partially confirmed, or blocked by missing evidence.
    
  • SKILL_CARD.md 1.4 KB
    Loading…

Comments (0)

Sign in to join the conversation.

No comments yet.

Reviews (0)

No reviews yet.