Skip to content

majiayu000/spellbook / skills/vibeguard

vibeguard

Verified

Lightweight anti-hallucination workflow for task kickoff, review prioritization, and regression retrospectives. Use when the user asks for guardrails, task contracts, risk scoring, or review templates.

FileKindSize
SKILL.mdskill md2.8 KB
references/review-template.mdmarkdown1.1 KB
references/scoring-matrix.mdmarkdown1.9 KB
references/task-contract.yamlconfig1.5 KB

references/task-contract.yaml

View on GitHub
task_contract:
  required:
    - goal: "A concrete, testable objective"
    - source_of_truth: "The file, API, document, or database that defines reality"
    - acceptance_criteria: "At least one verifiable done condition"
    - scope: "Files, modules, or systems allowed to change"

  forbidden:
    - "Start coding before the target behavior is clear"
    - "Invent fields, APIs, or data sources"
    - "Create a parallel implementation without searching first"
    - "Mark work done without verification evidence"

  warning_signs:
    no_search_before_create:
      severity: high
      description: "A new file, class, or helper is being added before checking for an existing implementation"
      action: "Search first and reuse or extend the current source of truth"

    weak_acceptance:
      severity: high
      description: "The task has no concrete done-when statement"
      action: "Rewrite the request as a testable outcome before coding"

    oversized_step:
      severity: medium
      description: "One step is too large to verify safely"
      action: "Split the change into smaller checkpoints with separate verification"

    missing_spec_for_multi_file_change:
      severity: medium
      description: "A multi-file change is starting without a design note or plan"
      action: "Write a short spec, dependency map, or execution plan first"

review_prompts:
  - "What exactly is being changed?"
  - "Where is the source of truth?"
  - "What proves the change is correct?"
  - "What is explicitly out of scope?"
planning-and-task-breakdownaddyosmani/agent-skillsBreaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
screenpipethedotmack/claude-memSearch screen and audio history recorded by Screenpipe and optionally save selected context into claude-mem. Use when the user asks to recall desktop activity, a meeting, or something seen in another app through Screenpipe.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
contact-widgetnexu-io/open-designSelf-contained floating chat widget with welcome screen, social links, meeting button, and message input. Single HTML file, zero dependencies.Stars: 108kUpdated Claude CodeCodexVerifiedInstall
meeting-notesnexu-io/open-designMeeting notes page — title bar with attendees, agenda checklist, decisions block, action items table with owners + dates, and a "next meeting" footer. Use when the brief mentions "meeting notes", "minutes", "1:1 notes", "all-hands recap", or "会议纪要".Stars: 108kUpdated Claude CodeCodexVerifiedInstall