plugins/development/skills/wrapup/SKILL.md
Performs end-of-session reflection by saving context to ACTIVE_CONTEXT.md, checking for mistakes to log, extracting lessons learned, and identifying documentation updates needed. Use after merging to development, when user says "wrapup", "zabal to", or at the end of a development session. NOT for saving context mid-session (use session-context) or for standalone mistake logging.
npx skillsauth add petrogurcak/skills wrapupInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Session reflection and context save after merge.
Announce: "Merge hotovy. Zavíram session."
Go through all 4 checks proactively:
Update .claude/ACTIVE_CONTEXT.md:
Review the session for correction patterns:
If yes → write entry to .claude/mistakes.md:
## [YYYY-MM-DD]: [Brief description]
**Co se stalo:** [What went wrong]
**Proc:** [Root cause]
**Oprava:** [How it was fixed]
**Pouceni:** [What to do differently]
**Tags:** #relevant #tags
If no → report "Zadne chyby."
Check if session produced reusable knowledge:
If yes → propose /learn entry or note in project memory.
If no → report "Zadne nove pouceni."
Check if the change affects any documentation:
.claude/DECISIONS.mdIf yes → propose specific update with file path and content.
If no → report "Dokumentace OK."
Zobraz kolik session spotřebovala a kde ušetřit — kritické pro Max plan rate limit tracking.
Aktuální 5h rolling block (Max limit window):
npx -y ccusage@latest blocks --active --json
Parsuj: costUSD, projection.totalCost, projection.remainingMinutes, models[], tokenCounts.*.
Týdenní přehled:
npx -y ccusage@latest weekly --json
Vezmi poslední týden: totalCost, identifikuj top den.
Top 3 token-heavy aktivity z transcriptu — projeď session a najdi co cerpalo nejvíc:
Read (soubory >1000 řádků)WebSearch / WebFetch chainsAgent spawns (Explore, general-purpose, research)Grep sweepy s output_mode: content a velkým head_limitSavings posouzení — pro každou heavy aktivitu rozhodni kam patří:
claude-glm (GLM Coding Plan, subscription covered) — bulk research, grep sweepy, mechanický refactor, long autonomous běhyclaude-kimi pay-per-token ($0,95 / $4 per 1M) — fallback nebo tasks s 262K context / image inputOutput format:
Usage:
- 5h block: $X.XX z projected $Y.YY (Z% využito, zbývá N min)
- Týden: $total.TT (top den: YYYY-MM-DD $day.DD)
- Modely: opus-4-7 $A.AA · glm-5.1 $B.BB · haiku $C.CC
- Heavy top 3: [1] <aktivita> ~Xk tokens [2] <aktivita> ~Yk [3] <aktivita> ~Zk
Savings tip: <1-2 konkrétní akční doporučení>
Příklad savings tip:
"Explore agent + 3 WebSearch chains (~$3.20) šlo přes claude-glm za $0. Příště: bulk research delegovat GLM, Opus jen na synthesis."
Fallback pokud ccusage selže (offline / network error):
Po vygenerování savings tipu:
Extrahuj 2-4 tags z tipu (lowercase, kebab-case, konkrétní). Příklady:
log-filter, server-side, bulk-handlinggh-cli, pr-investigation, scope-minimumtest-scope, tdd, pytestplan-reread, delegation, glmAppend do ~/.claude/savings-log.jsonl jako jeden řádek JSON:
{
"date": "YYYY-MM-DD",
"session_hint": "<projekt + topic>",
"tip": "<full savings tip text>",
"tags": ["tag1", "tag2", "tag3"],
"promoted": false,
"memory_files": []
}
Threshold check — spočítej frequency tagů napříč celým logem:
cat ~/.claude/savings-log.jsonl | jq -r '.tags[]' | sort | uniq -c | sort -rn
Auto-promotion rozhodnutí:
Pokud kandidát → proveď:
~/.claude/projects/-Users-petrogurcak-Projects/memory/feedback_<slug>.md s:
name, description, type: feedback, created, last_relevantMEMORY.md pod ### Execution efficiency (from wrapup savings tips) sekci (vytvoř pokud neexistuje)"promoted": true, "memory_files": ["feedback_<slug>.md"]Pokud pattern ≥3× a už existuje memory file → navrhni promotion do ~/.claude/CLAUDE.md Execution Efficiency sekce (hard rule, aplikuj vždy). Neprováděj automaticky — vždy user potvrdí.
Oznam v Present Summary (viz níže): kolik tipů logged, kolik promoted, kolik kandidátů na CLAUDE.md rule.
Anti-spam pravidla:
--foo flag u tohohle konkrétního commandu")Session wrap-up:
- Kontext: Ulozeno
- Mistakes: [zadne / zapsano: "brief description"]
- Lessons: [zadne / zapsano: "brief description"]
- Dokumentace: [beze zmen / aktualizovano: file.md]
- Usage: $X.XX session / $Y.YY 5h block (Z% limit)
└─ Savings: <1-liner tip>
└─ Log: +N tip(ů) → savings-log.jsonl · promoted: <count nebo "—">
└─ Promotion kandidát: <tag s freq ≥3 nebo "—"> → CLAUDE.md rule?
Hotovo. Neco dalsiho?
development
Builds a pre-launch social proof strategy through structured beta programs using D'Souza Brain Audit interviews. Use when launching new products/services and need compelling testimonials, planning a beta cohort, designing interview questions to harvest objection-busting social proof, improving video testimonials for landing pages, or designing case studies with metrics. Trigger phrases include "beta tester program for testimonials", "pre-launch social proof", "Brain Audit testimonial framework", "case study harvest", "reverse testimonial", "video testimonial mechanics", "social proof landing page", "sběr referencí", "beta tester program", "testimonial pro landing page", "social proof před launchem", "rozhovor s klientem", "case study sběr", "reference před spuštěním". NOT for ongoing case study production (use growth-hacking case-study approach), offer design (use offer-creation), or conversion optimization (use ux-optimization).
development
Use when planning a product launch and the product type is unclear or could be either generic (SaaS/app/physical) or info-product. Routes between marketing:launch-strategy (generic launches) and marketing:info-product-launch (courses, memberships, ebooks, cohorts, communities). Trigger phrases - "launch", "spuštění", "go-to-market", "product launch", "release strategy", "uvedení na trh", "launch plan", "spuštění produktu", "launch sequence", "launch strategy". Do NOT trigger when product type is already clear (use specific skill directly).
testing
Specialized 8-week launch cadence for info-products — online courses, cohort programs, memberships, communities, ebooks, masterminds. Combines Jeff Walker's Product Launch Formula (Seed/Internal/JV variants, PLC sequence, open-cart day-by-day) with Stu McLaren's membership mechanics (closed cart, Success Path) and Hormozi Grand Slam Offer stacking. Use when planning "launch online kurzu", "info-product launch", "PLF launch", "course launch", "membership launch", "cohort launch", "ebook launch", "open cart close cart", "8-week launch of online course", "beta cohort to launch sequence", "spuštění kurzu", "launch členské sekce", "open cart strategie". Differentiates from marketing:launch-strategy (generic SaaS/app launches) — info-product-specific. NOT for SaaS launches, physical products, or services.
development
Use when releasing an Expo/React Native mobile app to App Store and Google Play - covers eas submit, ASC "Submit for Review", Play promote Internal→Production, OTA update, and decoding common silent failures (Apple agreement expiry, missing English locale, Background Location declaration, web bundle failure on react-native-maps).