一键导入
flashcard
Use when recording project-specific mistakes, gotchas, or hard-won knowledge. Also activates proactively when a mistake pattern is detected during work.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Use when recording project-specific mistakes, gotchas, or hard-won knowledge. Also activates proactively when a mistake pattern is detected during work.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when you need to delegate implementation tasks to specialized agents (Developer, QA, Architect, DevOps, Scribe) as the Team Lead. Activates when orchestrating the Bounce phase — delegating stories, monitoring reports, merging completed work, consolidating sprint results, and generating product documentation. Uses git worktrees for story isolation. Works with Claude Code subagents natively; file-based handoffs for other tools.
Use when creating, modifying, or navigating V-Bounce Engine planning documents. Trigger on any request to create a charter, roadmap, epic, story, sprint plan, or risk registry — or when the user asks to update, refine, decompose, or transition documents between phases. Also trigger when the user asks about work status, backlog, what's next, what's blocked, or wants to plan/start a sprint. This skill manages the full document lifecycle from Charter through Sprint Planning and execution.
Use when the V-Bounce Engine framework needs to evolve based on accumulated agent feedback. Activates after sprint retros, when recurring friction patterns emerge, or when the user explicitly asks to improve the framework. Reads Process Feedback from sprint reports, analyzes FLASHCARDS.md for automation candidates, identifies patterns, proposes specific changes to templates, skills, brain files, scripts, and agent configs with impact levels, and applies approved changes. This is the system's self-improvement loop.
Use when you need to understand document relationships, check what's affected by a change, find blocked documents, or assess the state of planning documents. Provides structured awareness of the full product document graph without reading every file. Auto-loaded during planning sessions.
**Codebase Cleanliness Standard**: Enforces clean file organization in any codebase. Before creating ANY file, classify it by intent—deliverables go to the project tree, working artifacts go to `/temporary/`. Before committing, review `git diff` to catch misplaced files. Use this skill whenever creating, moving, or committing files. Works with all languages and frameworks. The `/temporary/` folder is git-ignored so working artifacts never get merged. ALWAYS consult this skill when writing files to the repo—it prevents clutter from debug scripts, scratch analysis, throwaway tests, and other AI working artifacts from polluting the codebase.
A code quality and architecture review skill designed for AI-generated (vibe-coded) codebases. Use this skill whenever the user asks to review, audit, analyze, or assess code quality of a project — especially one built with AI agents. Trigger on phrases like 'review my code', 'check code quality', 'is my architecture solid', 'run a health check', 'audit my codebase', 'check for tech debt', 'analyze my project', or any mention of code review for vibe-coded or agent-generated projects. Also trigger when the user asks about architectural consistency, error handling quality, duplication, coupling, dependency health, or test quality. This skill runs in phases — from quick PR-level diffs to full codebase scans — and produces structured reports a non-coder can understand.
| name | flashcard |
| description | Use when recording project-specific mistakes, gotchas, or hard-won knowledge. Also activates proactively when a mistake pattern is detected during work. |
Captures project-specific mistakes and rules into FLASHCARDS.md so they are never repeated. YOU MUST read FLASHCARDS.md before modifying code in any session.
Core principle: Every mistake is an investment — but only if you record it.
/flashcard OR /flashcard [description] OR proactively when a mistake or gotcha is detected during work.
When using this skill, state: "Recording a flashcard."
This is NOT just a command — it is a standing directive:
FLASHCARDS.md at the project root. Treat recorded rules as hard constraints.Lessons MUST be recorded as soon as the story that produced them is merged — not deferred to sprint close. Context decays fast.
Flow:
lessons_flagged field)lessons_flagged from Dev and QA reportsWhy this matters: A lesson recorded 5 minutes after the problem is specific and actionable. A lesson recorded 3 days later at sprint close is vague and often forgotten.
/lesson CommandIf the user provides a description (/lesson [description]), use it. Otherwise:
WAIT for user input if context is unclear.
Use this exact format — no deviations:
### [YYYY-MM-DD] Short descriptive title
**What happened:** One or two sentences describing the problem or mistake.
**Rule:** The actionable rule to follow going forward. Write as an imperative.
Rules for formatting:
FLASHCARDS.md at the project root# FlashcardsFLASHCARDS.md lives at the project root. Flat, chronological, no categories.
# Flashcards
### [2026-02-18] RLS policies break cascade deletes
**What happened:** Tried cascade delete on projects table, silently failed due to RLS.
**Rule:** Always use soft deletes with RLS. Never cascade.
### [2026-02-15] Vercel preview URLs break CORS
**What happened:** OAuth failed on every preview deploy because preview URLs weren't in the CORS allowlist.
**Rule:** Use wildcard pattern for Vercel preview branch origins in CORS config.
FLASHCARDS.md before modifying project code. No exceptions.Lessons that have been proven effective across 3+ sprints become permanent agent config rules.
A lesson is a graduation candidate when:
A lesson qualifies for accelerated graduation (after 1 sprint instead of 3) when:
Accelerated candidates are flagged by suggest_improvements.mjs with impact level P1. The human still approves — the only difference is the 3-sprint waiting period is waived.
.vbounce/scripts/suggest_improvements.mjs flags graduation candidates in improvement suggestions.claude/agents/developer.md, etc.)FLASHCARDS.md with a note: [Graduated to {agent} config on {date}].vbounce/improvement-log.md under "Applied"FLASHCARDS.md is a staging area, not a permanent rule store. Lessons that graduate become enforced constraints in the agent's core instructions — they can't be forgotten or overlooked. Lessons that stay in FLASHCARDS.md are read on every session but are softer guidance. Keep FLASHCARDS.md lean — stale lessons dilute the signal.