# Growth Retrospective > Personal growth gap retrospective. 5-category sweep ranked by decision-velocity signals; each top gap launched with a Deep Research / NotebookLM prompt + 4 multiple-choice questions grounded in your own codebase (your files + ADRs are the ground truth, not the AI's knowledge). Step 0 Action Audit refuses new items if prior level-up actions are stuck — designed so items leave the list. Use when slash commands `/retrospective`, `/reflect`, `/growth-review`, or keywords like "what should I learn", "where am I weak", "am I improving", "review my growth". Also proactive after long sessions (≥30 substantive turns) or milestones (PR merged, ADR landed, sprint closed). Cross-agent compatible — Claude Code, Codex CLI, Antigravity, Cursor, VS Code, same SKILL.md format, agent-specific storage paths. - Skill: `youngapple/growth-retrospective` (Agent Skill, multi-file: 11 files) - Install (CLI): `npx skillmds add youngapple/growth-retrospective` - Raw SKILL.md: https://api.skillmd.com/api/skills/youngapple/growth-retrospective/raw - Safety review: pending - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Marketing & Growth - License: MIT - Author: YoungApple (https://skillmd.com/u/youngapple) - Updated: 2026-09-10 - Page: https://skillmd.com/skills/youngapple/growth-retrospective --- # Growth Retrospective A retrospective skill designed so items leave the list. Captures personal growth gaps across 5 categories, ranks by decision-velocity signals, and **launches** learning — not just lists it. North star: **decision velocity × decision quality** trending up over months. A log that only grows is failure mode. ## When to invoke - Slash commands: `/retrospective`, `/reflect`, `/growth-review` - Keywords (any language): "what should I learn", "where am I weak", "review my growth", "am I improving", "我的 gap 在哪", "需要学什么" - Proactive: after ≥30 substantive turns in a session, or after PR merged / ADR landed / sprint close. Ask once at a natural stopping point; never interrupt mid-task. - First-time setup: full baseline scan across the whole repo + chat history. If multiple triggers fire together, run ONCE — never double up. ## Core loop — 5 phases | # | Phase | What | |---|---|---| | 0 | **AUDIT** | Did prior level-up actions actually ship? Filesystem check before adding new gaps. | | 1 | **SWEEP** | Visit all 5 categories explicitly. Missing one is a defect, even if empty. | | 2 | **RANK** | Tier each gap by decision-velocity signals. | | 3 | **ANCHOR** | Ground each top gap to a real `file:line` / ADR / commit in the user's repo. | | 4 | **LAUNCH** | Per top gap: research prompt + anchored MCQ. | All 5 phases are mandatory. Skipping AUDIT turns the log into a hoarder list. Skipping SWEEP biases toward domain knowledge. Skipping ANCHOR weakens verification. Skipping LAUNCH leaves gaps decorating the log forever. --- ## Phase 0: AUDIT — the forcing function (anti-busywork) Before adding ANY new gap, read the prior detailed reference and check: 1. List every prior `How to level up` action across all Tier 1 + Tier 2 items. 2. For each, check the filesystem: does the named artifact exist? (cheatsheet, script, diagram, test, alert, ADR entry) 3. Decision rule: - **0 completed AND ≥3 pending >14 days** → REFUSE to add new items. Offer: - A) Commit to one pending action (mark it active, no new scan) - B) Explicitly drop pending actions (note in detailed ref, no new scan) - C) Override with a documented reason (new scan proceeds; reason logged) - **Anything else** → proceed. The refusal path is the feature. The skill's value lies in saying "no more new items until you close the old ones." --- ## Phase 1: SWEEP — visit all 5 categories explicitly | # | Category | Visible in | |---|---|---| | 1 | Domain knowledge | ADRs, PR review rounds, repeated concept lookups | | 2 | Human skills | Chat patterns over weeks (negotiation, written precision, asking) | | 3 | Work habits | Git history (ADR discipline, doc follow-through, commit hygiene) | | 4 | Meta-cognition | Cross-domain patterns (how you decide, when you ask vs push) | | 5 | Productivity / focus | Commit timestamps, session lengths, interruption signals | Most users default to (1). The compounding lives in (4) and (5). **Accelerants** (optional, fall back to manual `git log` / session-jsonl reading if scripts can't run): - `bash scripts/scan_git_signals.sh [days]` - `python3 scripts/scan_chat_signals.py --project-id= --topics=` (Note: `--project-id=` requires the `=` form — IDs start with `-` and argparse mis-parses positional `-` strings as flags.) --- ## Phase 2: RANK — tier each gap by signals See `references/signal_taxonomy.md` for the full taxonomy. - **Tier 1 — actively slowing decisions.** Strong signals: ≥3 review rounds on same topic, reversed ADRs, evidence-falsified challenges, repeated "what does X mean" in same week. - **Tier 2 — knowledge gap, not yet blocking.** Medium signals: ADR landed after challenge, decided-but-not-automatic, looked up rather than recalled. - **Tier 3 — settled.** Decisions stick. Demote here when the graduation marker is met. Graduating off the list is the success state. **Mandatory demotion check:** every pass, walk all Tier 1 + Tier 2 items and ask whether the graduation marker has quietly been met. Demote if yes. Items don't graduate on their own. --- ## Phase 3: ANCHOR — ground each top gap to user's real artifacts For each Tier 1 + Tier 2 gap (max 3 surfaced per retro): 1. Grep the user's repo for related symbols, error messages, ADR references. 2. Find a concrete `:` / `ADR-N` / `` the user can open in under 2 minutes. 3. If no anchor found in ≤2 min of searching, mark the gap **abstract** and warn — Phase 4 MCQ will be weaker. Anchoring is what makes the next phase work. AI-native skills can read the user's code; traditional learning apps can't. This is the unique leverage point. --- ## Phase 4: LAUNCH — emit a learning launchpad per top gap For each top gap (max 3), produce TWO artifacts: ### (a) Research prompt (copy-paste to Deep Research / NotebookLM / ChatGPT) ``` I'm building [project, 1 line]. I need to understand [gap concept] in the context of [anchor — : or ADR-N]. Questions: 1. When does this matter in practice? 2. What are the failure modes I should worry about? 3. Real-world examples in [user's stack]? 4. Common misconceptions? Output: deep-dive report with citations to primary sources (docs, RFCs, post-mortems). ``` The prompt is portable. The user picks the external tool. The skill doesn't bind to one vendor. ### (b) Anchored MCQ (4 questions, 2-3 min total) Generate 4 multiple-choice questions verifying the user learned the concept. **Ground truth comes from the user's own artifacts, not the AI's knowledge** — this is the defense against AI-hallucinated answers. Use a mix of question forms: | Form | Example | Ground truth | |---|---|---| | **State** | "Your `:` uses isolation level X. Which?" | `grep` the file | | **Consequence** | "If ADR-N's choice were reversed, what breaks first?" | ADR-N body | | **Compare** | "Run A vs Run B — which is more likely to regress on category X?" | Eval output | | **Predict-then-reveal** | "Before I show you ADR-M's actual rationale, predict: (a)(b)(c)(d)" | ADR-M body | The predict-then-reveal form is highest-retention (prediction-error feedback in learning science). Use it for ≥1 of the 4 questions when a relevant ADR or commit rationale exists. **Constraints**: - Each question ≤30 sec to answer. - 4 questions = 2-3 min total. - Answer key MUST cite the ground-truth artifact: `Correct: (b). See :`. - Skip MCQ for **taste gaps** (design opinion, product judgment). Use Explain-back instead: "Describe in 5 sentences why X over Y." - Skip MCQ for **muscle memory** (vim, shell fluency). Verification mechanism doesn't solve those — only practice does. The skill's job is to make starting trivially easy. Whether the user follows through is their choice. --- ## Storage — two layers, both mandatory | Layer | Path (Claude Code) | Lifecycle | |---|---|---| | **Short index** | `~/.claude/projects//memory/learning_domains.md` | Auto-loaded each session; ≤80 lines / 4 KB; one bullet per active item | | **Detailed ref** | `/docs/learning-domains.md` | Grep'd on demand; 5 fields per item: `Signal · Concepts · Status · How-to-level-up · Graduation marker` | Cross-agent short-index paths (write to whichever the host agent auto-loads): | Host | Short-index path | |---|---| | Claude Code | `~/.claude/projects//memory/learning_domains.md` | | Codex CLI | Append `## Learning domains` section to repo's `AGENTS.md` | | Antigravity | Append to repo's `.agents/agents.md` | | Unknown | `/.growth-log/short-index.md` | The detailed reference is repo content (same path everywhere); the short index is agent-state. **Both layers MUST stay in sync.** Adding to one only is a defect. When an item graduates, simplify in detailed ref + remove from index. Templates: `assets/templates/short_index.md`, `assets/templates/detailed_reference.md`. --- ## Output to the user (chat reply) After running, the user sees ONLY: 1. **Phase 0 verdict** — if AUDIT refused, show this FIRST and stop. 2. One-paragraph summary: graduations / new items / stuck items. 3. File paths: short index + detailed ref. 4. Top 1-3 launchpads (research prompt + MCQ block) — inline, ready to copy-paste. Do NOT dump the full updated detailed reference. It's a file; the user opens it. ### Communication discipline (HARD) Before sending the chat reply, scan `~/.claude/projects//memory/` for files matching `user_language_preference.md` or `feedback_*.md`. Apply their checklists. For Chinese-primary users: first sentence + section headers in Chinese, no fragment-by-code-switch, jargon glossed on first mention. The skill MUST NOT cost the user the communication discipline they trained. --- ## Incremental vs full mode - **Incremental** (default for milestones / session-end): scan only the delta since last review. ≤5 min. - **Full** (explicit slash command, weekly cadence, first-time): full repo + chat-history scan. ≤20 min. Infer from context if unspecified. Explicit slash command → full unless user says otherwise; everything else → incremental. --- ## Anti-patterns | Don't | Why | |---|---| | Track expertise | The log is for what's still hard, not what's mastered | | Pad with vague items ("be a better engineer") | Vague = unverifiable = never graduates | | Generate inspirational content | Terse and signal-dense only | | Re-run on every session | Proactive trigger fires AT MOST once per session | | Skip Phase 0 | Bypassing the forcing function turns the log into a hoarder list | | Skip Phase 4 LAUNCH | Identifying gaps without launching is decoration | | Generic MCQ ("explain X") | MCQ must be anchored to user's own files | | Hand-wave the MCQ ground truth | Every correct answer cites a verifiable artifact | | Re-introduce a graduated item as new | Check the log first; escalate, don't duplicate | | Block on missing scripts | Scripts are accelerants; manual extraction always works | --- ## Cross-references - `references/signal_taxonomy.md` — full decision-velocity signal list - `references/growth_categories.md` — deeper guidance on the 5 categories - `references/learning_journey.md` — level-up actions + graduation markers - `references/examples.md` — worked example from `squishy-platypus` - `assets/templates/short_index.md` + `assets/templates/detailed_reference.md` - `scripts/scan_git_signals.sh`, `scripts/scan_chat_signals.py`, `scripts/render_storage.py` If any reference file is missing for the user's project, use templates as guidance and produce the layers by hand. The skill works without the scripts.