Skip to main content
Jeden Skill in Manus ausführen
mit einem Klick
GitHub-Repository

agentic-engineering-course-public

agentic-engineering-course-public enthält 38 gesammelte Skills von genkovich, mit Repository-Berufsabdeckung und Skill-Detailseiten auf SkillsMP.

gesammelte Skills
38
Stars
4
aktualisiert
2026-07-15
Forks
5
Berufsabdeckung
5 Berufskategorien · 100% klassifiziert
Repository-Explorer

Skills in diesem Repository

prepare-vps-deploy
Softwareentwickler

Prepare and verify the repository files required to deploy this Next.js project to one Ubuntu VPS with Docker Compose and a GitHub self-hosted runner. Use when creating or reviewing Dockerfile.production, compose.vps.yml, the health route, persistent SQLite storage, or deploy-vps.yml before a VPS release.

2026-07-15
bump-version
Softwareentwickler

Decide and apply the next semver for a release. Triggers on "bump the version", "what version is next", "/bump-version", "простав версію", "яка наступна версія". Reads the next number from the deterministic scripts/next-version.sh (NEVER guesses it), explains in plain language WHICH part moved and why (e.g. "MINOR, because a feat landed: filter notes by tag"), edits the version line in pyproject.toml, and proposes the matching `git tag vX.Y.Z`. The number is decided by a rule; the agent explains and applies — the human confirms the tag. Never pushes a tag itself.

2026-07-06
check-docs-drift
Softwareentwickler

Find and fix where docs/api.md fell behind the code. Triggers on "check for doc drift", "is the API doc up to date", "/check-docs-drift", "перевір дрейф документації", "docs drift". Runs the deterministic scripts/check-docs-drift.py (parses the public NoteBook methods with ast, compares to what docs/api.md documents — NO LLM finds the gap), reports the drift, and PROPOSES the doc edit that closes it. The gap is found by parsing; the agent writes the missing documentation in the doc's existing style. Edits only docs/; never commits or pushes.

2026-07-06
codify-rule
Softwareentwickler

Turn a repeated mistake into a durable, path-scoped rule in .claude/rules/. Triggers on "codify this rule", "make a rule so this doesn't happen again", "/codify-rule", "закодифікуй правило", "петля помилка-правило". Use only when the same mistake showed up TWICE (twice = a pattern, once = noise). Names the wrong shape and the right shape with one short example of each, scopes the rule to the paths that matter, and leaves the edit in the working tree so the next agent on this repo inherits the fix. Edits only .claude/rules/; never commits or pushes.

2026-07-06
curate-changelog
Softwareentwickler

Curate the [Unreleased] section of docs/CHANGELOG.md from git history. Triggers on "curate the changelog", "update the changelog", "/curate-changelog", "онови changelog", "курування журналу". Reads git log since the last tag, FILTERS out what a reader never feels (test/ci/chore/merge), GROUPS commits about one feature into one bullet, and REWRITES technical subjects into plain language under the six Keep a Changelog categories (Added/Changed/Deprecated/ Removed/Fixed/Security). The result must have FEWER lines than the raw log — that is the proof of curation, not a dump. Leaves the edit in the working tree; never commits or pushes.

2026-07-06
release-notes
Softwareentwickler

Draft partner-facing release notes from the same commits as the changelog. Triggers on "draft release notes", "release notes for the next version", "/release-notes", "нотатки релізу", "release notes партнеру". Same input as curate-changelog, different audience and tone: a short, benefit-first narrative for a user or partner — "what's new for you" — not a categorized engineer's list. Keeps only changes a user actually notices; drops internal refactors and tooling. Prints the notes to the chat to seed a GitHub Release description or a partner announcement. Writes no file.

2026-07-06
release
Softwareentwickler

Run the whole local release pipeline in order, stopping at each human gate. Triggers on "run the release", "prepare a release", "/release", "зроби реліз", "підготуй реліз". Drives the four stages back to back — bump-version → curate-changelog → release-notes → check-docs-drift — pausing after each so the human reviews the diff before the next stage. Prepares everything in the working tree; never tags, commits, or pushes. The tag push (which fires CI) stays the human's action.

2026-07-06
verify-gate
Softwareentwickler

Реалізувати задачу за ДЕТЕРМІНОВАНИМ гейтом — ганяти `node --test` і НЕ казати DONE, поки тести червоні; вставити вивід як доказ. Порт 7.6 verify-gate під стек node:test.

2026-07-04
tdd
Softwareentwickler

Провести задачу через RED → GREEN → REFACTOR з рівно 3 атомарними комітами (test/feat/refactor) на стеку node:test. Порт оркестратора 7.7. Tests — незмінний контракт після RED.

2026-07-04
code-reviewer-subagent
Softwarequalitätssicherungsanalysten und -tester

Піднімає рев'юера у ЧИСТОМУ контексті, що бачить ЛИШЕ git diff + AC PR, і повертає дискретний вердикт ACCEPT/WARN/PARTIAL/REJECT з обґрунтуванням. Тригери — '/code-reviewer-subagent', 'дай рев'юера в чистому контексті', 'субагент має оцінити diff проти AC', 'review this diff with fresh eyes'. НЕ вбудований '/code-review'.

2026-06-25
welcomer
Projektmanagementspezialisten

Greet a user starting a new project and walk them through a minimal onboarding checklist. Use when the user opens a fresh repo, says "let's start", "новий проєкт", "kickoff", or asks how to begin. Use even if the user does not mention onboarding or welcome by name.

2026-06-09
welcomer
Projektmanagementspezialisten

Greet a user starting a new project and walk them through a minimal onboarding checklist. Use when the user opens a fresh repo, says "let's start", "новий проєкт", "kickoff", or asks how to begin. Use even if the user does not mention onboarding or welcome by name.

2026-06-09
ralph-prep
Softwareentwickler

Generate a PROMPT.md scaffold for the Ralph loop from a story file. Triggers on '/ralph-prep <story-id>' (e.g. '/ralph-prep story-t1'), or when the user asks to 'підготуй PROMPT для Ralph', 'згенеруй PROMPT.md з story', 'prepare Ralph PROMPT from <story>'. Reads tasks/<story-id>.md + CLAUDE.md and writes the 3-section prompt to ./PROMPT.md.

2026-06-02
code-reviewer-subagent
Softwarequalitätssicherungsanalysten und -tester

Spin up a clean-context subagent reviewer that sees ONLY the git diff plus a story's AC, and returns a discrete ACCEPT/WARN/PARTIAL/REJECT verdict with rationale. Triggers on '/code-review-subagent <story-id>' (e.g. '/code-review-subagent story-27'). NOT the built-in '/code-review'. Use when the user asks to 'дай рев'юера в чистому контексті', 'субагент має оцінити diff проти AC', 'review this diff with fresh eyes'.

2026-06-02
screenshot-diff
Softwarequalitätssicherungsanalysten und -tester

Compare a rendered component against a design baseline pixel by pixel. Triggers on '/screenshot-diff <story-id>' (e.g. '/screenshot-diff story-26'), or when the user asks to 'звір з Figma-еталоном', 'screenshot diff проти baseline', 'перевір верстку по скриншоту'. Reads baselines/<feature>.png, renders the component on :5173, snapshots to tmp/, computes the diff, returns pass/fail plus a diff image.

2026-06-02
verify-gate
Softwarequalitätssicherungsanalysten und -tester

Implement a story behind a DETERMINISTIC gate — run `tsc --noEmit` + `vitest run`, and do NOT say DONE until both are green, pasting the command output as evidence. Triggers on '/verify-gate <story-id>' (e.g. '/verify-gate story-28'), or 'детермінований гейт', 'доказ перед DONE', 'не кажи готово поки тести червоні'. Reads tasks/<story-id>.md for AC, iterates on the implementation using the failing test output as feedback, refuses completion while red.

2026-06-02
verify-ui
Softwarequalitätssicherungsanalysten und -tester

Verify a story's acceptance criteria through a LIVE browser run via playwright-cli (CLI-first, token-efficient). Triggers on '/verify-ui <story-id>' (e.g. '/verify-ui story-26'), or 'перевір UI у браузері', 'прокликай story через Playwright', 'розробка через Playwright'. Reads tasks/<story-id>.md, opens :5173, drives the scenario, reads real page state (DOM + computed styles), returns a per-AC pass/fail with a screenshot artefact. Does NOT fix code — only renders a verdict.

2026-06-02
tdd
Softwareentwickler

TDD-orchestrator що проганяє повний Red-Green-Refactor цикл через 3 ізольовані agents. Використовуй коли користувач каже `/tdd <story-id>` (наприклад `/tdd story-24-sm2`), або «прогон TDD циклу на story X», «запусти TDD pipeline», «зроби RGR на цій story». Послідовно викликає tdd-test-writer → tdd-implementer → tdd-refactorer через Agent tool, з automatic bash-gates між фазами. Підтримує опційний flag `--review-tests` для зупинки після RED і human review.

2026-06-02
api-forge
Softwareentwickler

Use when user wants to generate API contract (OpenAPI 3.1 / GraphQL / async events) per SDLC stage 10 (api-contracts) protocol. Triggers on "api-forge for {slug}", "API for {slug}", "openapi for {feature}", "GraphQL schema", "AsyncAPI", "events for {feature}", "stage 10 for {slug}", "/sdlc-api-forge {slug}". Output: docs/features/{slug}/contracts/openapi.yaml + docs/features/{slug}/contracts/api-sync-report.md + contracts/events.md (if async). Hard gate: docs/features/{slug}/PRD.md (stage 03). data-model.md (stage 08) is recommended but optional — without it the skill runs in scenario B (PRD + sequences only) and fills `unresolved_origins` in the sync report. Renamed from `define-api` in v3.3.0; the legacy name is kept as a deprecation wrapper.

2026-05-25
architecture-design
Softwareentwickler

Use when user wants to produce a Software Architecture Document (SAD) with Arc42 12 sections + ADRs for a feature, after PRD is closed. Triggers on "architecture for {slug}", "design architecture for {feature}", "SAD for {slug}", "arc42 for {slug}", "stage 04-05 for {slug}", "C4 context+container for {slug}", "/sdlc-architecture-design {slug}", "/sdlc-arc {slug}". One skill replaces draft-architecture + write-arc42 + draw-c4 + propose-adr. Drafts §1-§12 in-memory, then per-section batch validates via AskUserQuestion (4-state machine: Approve / Edit / Save as Open Question / Drop), spawns ADRs only on blast-radius gate (irreversible / multi-module / has legitimate alternatives), writes the section + commits atomically. Clean-context critic on the post-Socratic SAD before the finalization commit. C4 levels 1-2 inline (Context in §3, Container in §5); no L3/L4. Brownfield: dispatches Explore subagent to map the repo before drafting. Prerequisite: docs/features/{slug}/PRD.md — hard refuse if missing. GATE stage 🚪

2026-05-25
break-tasks
Softwareentwickler

Use when user wants to break implementation into atomic tasks per SDLC stage 13 (task-breakdown) protocol. Triggers on "task breakdown for {slug}", "break down tasks for {feature}", "tasks for {slug}", "sprint plan for {feature}", "stage 13 for {slug}", "/sdlc-break-tasks {slug}". Output: docs/features/{slug}/tasks/_epic.md + tasks/tracker.md + tasks/<task>.md with atomic tasks ≤1 day each + dependency graph + DoD per task. Prerequisite: docs/features/{slug}/PRD.md + sad.md + ≥1 Accepted ADR — hard refuse if missing. Stories LINK to upstream artefacts (PRD AC §, SAD §6, data-model, openapi), they do not duplicate them.

2026-05-25
classify-size
Softwareentwickler

Use to classify a feature into XS/S/M/L/XL and write docs/features/{slug}/.size so phase wrappers can skip arc42/c4-component/deployment/test-plan/ security-review per MVP-vs-Full matrix. Triggers on "classify size", "feature size", "XS or M for {slug}?", "/sdlc-classify-size {slug}", "size {slug}". Asks 4 AskUserQuestion (PR count / time / new module-API- migration / breaking changes) → maps to XS/S/M/L/XL → writes a one-line file docs/features/{slug}/.size. Syncs with PRD.md frontmatter `feature_size:` if present (warns on conflict). Output: docs/features/{slug}/.size + (opt.) PRD frontmatter update.

2026-05-25
complete-sequence-diagrams
Softwareentwickler

Use when the user wants to audit a SAD against PRD user stories and add sequence diagrams for every uncovered use case (not just the 3-5 critical ones). Triggers on "complete sequences for {slug}", "audit sequence coverage", "missing sequences", "sequence per US", "/sdlc-complete-sequences {slug}". Iterative: one use case at a time with user confirm. Output: Mermaid sequenceDiagram blocks appended/updated inline in docs/features/{slug}/sad.md §6 (Runtime view). Also handles single-flow ad-hoc draws (superset of legacy draw-sequence skill). Prerequisites: docs/features/{slug}/PRD.md (stage 03) + docs/features/{slug}/sad.md (stage 04) — hard refuse if missing.

2026-05-25
decide-adr
Softwareentwickler

Use for standalone / post-hoc ADR documentation outside the synchronous architecture-design pass. Triggers on "ADR for {decision}", "adr for {slug}", "decide on ADR", "lock in decision", "MADR for {topic}", "stage 10 for {slug}", "/sdlc-decide-adr {slug} {title}". Use this when a decision was made earlier (in code, in a chat, on a whiteboard) and only needs to be recorded — for synchronous design-time decisions use sdlc:architecture-design (which spawns ADRs inline). Output: docs/features/{slug}/adr/NNNN-{title}.md with context / options / decision / consequences / status. Prerequisite: docs/features/{slug}/PRD.md + sad.md (stages 03 + 04-05) — hard refuse if missing. GATE stage 🚪 — ADR merged with status Accepted before downstream stages (break-tasks / prep-context / ship).

2026-05-25
fix-term
Softwareentwickler

Use when a domain term shows up in conversation (interview, PRD, ADR) and you want to fix its meaning in CONTEXT.md glossary before it drifts. Triggers on "fix term X", "what is X in our domain?", "add to CONTEXT", "fix term {X}", "/sdlc-fix-term {term}". Lazy-bootstraps CONTEXT.md (single-context or multi-context via CONTEXT-MAP.md), asks for canonical definition + NOT-cross-reference, checks conflict with existing entries, appends a line to ## Glossary. Output: edited CONTEXT.md (created lazily if missing). Skip for generic tech terms (HTTP, database) — that's not a domain glossary. Primary invocation context: `sdlc:interview` triggers this skill automatically at phase 3 (Glossary checkpoint) when its glossary-matching spots a new domain term in a user answer; standalone invocation is for manual edits or for retro-fitting an existing glossary.

2026-05-25
generate-data-model
Datenbankarchitekten

Use when the user wants to design the data model AND generate the actual forward + rollback migrations in one pass. Triggers on "data model for {slug}", "schema for {feature}", "generate migrations for {slug}", "DB design + migration", "stage 8+9 for {slug}", "/sdlc-generate-data-model {slug}". Default mode is greenfield-first: reads PRD §4 + SAD §6.4 ER + sequence diagrams + (optional) Go domain structs and produces docs/features/{slug}/data-model.md + migrations/*.up.sql + migrations/*.down.sql + a markdown report. Supersedes legacy design-db + plan-migration skills. Prerequisites: docs/features/{slug}/PRD.md (stage 03) + docs/features/{slug}/sad.md (stage 04) — hard refuse if missing.

2026-05-25
interview
Softwareentwickler

Consolidated 14-phase ideation skill — Socratic interview, autonomous competitive research, 3 strategic approaches via parallel sub-agents, multi-perspective review (Engineer/Executive/UX), devil's advocate, Claude-proposed RICE/Feasibility. Single entry-point for ideation phase. Produces idea-brief.md (15 sections, ≤5 pages). Triggers on "raw idea", "capture an idea", "interview a feature X", "brief for X", "new feature X", "idea brief", "intake feature X", "start new feature", "ideation for {slug}", "/sdlc-interview {slug}". Replaces the prior intake + brainstorm + interview trio. ADRs are no longer part of this skill — they are spawned inline by the architecture-design skill at gate 04-05. Not to be confused with the global `interview` skill (stress-testing ideas) — this one is bound to SDLC ideation phase and writes an artifact into docs/features/.

2026-05-25
plan-tests
Softwarequalitätssicherungsanalysten und -tester

Use when user wants to write a test plan per SDLC stage 15 (test-plan) protocol. Triggers on "test plan for {slug}", "test plan for {feature}", "what we test", "test strategy", "stage 15 for {slug}", "/sdlc-plan-tests {slug}". Output: docs/features/{slug}/test-plan.md with unit/integration/E2E breakdown, AC mapping (each AC → ≥1 test), test data strategy. Prerequisite: docs/features/{slug}/PRD.md (stage 03, for AC mapping) — hard refuse if missing.

2026-05-25
write-prd
Softwareentwickler

Use when drafting a PRD for SDLC stage 03 with idea-brief.md + CONTEXT.md present. Reads required inputs, asks via AskUserQuestion which optional channels to use (--reference module code, MCP-Atlassian, RAG, docs), drafts from ./templates/PRD-template.md, Socratically validates per item, then runs a clean-context critic on the edits-log, then writes the file. Hard refuse if idea-brief.md or CONTEXT.md missing. Triggers: "/sdlc-write-prd {slug}", "write PRD for {slug}", "draft PRD for {slug}", "PRD для {slug}", "write spec for {slug}", "product requirements for {slug}". Output: docs/features/{slug}/PRD.md.

2026-05-25
legacy-critic
Softwarequalitätssicherungsanalysten und -tester

Use when running Step 2.2 of legacy refactoring protocol — independent critic subagent verifies LEGACY/<module>.md against random source files, claim-by-claim. Triggers on '/legacy-critic <module>' or when user asks for 'перевірка протиріч', 'critic pass', 'verify legacy summary'.

2026-05-16
legacy-cutover
Softwareentwickler

Use when running Step 4 of legacy refactoring protocol — writes one use case worth of new code in internal/<module>/, switches DI, ensures pytest stays green, and appends to REFACTOR_LOG.md. Triggers on '/legacy-cutover <module> --usecase=<name>' or when user asks to 'переписати register', 'cutover register', 'переписування у нову папку'.

2026-05-16
legacy-extract
Softwareentwickler

Use when running Step 2 of legacy refactoring protocol — extracts a 6-section summary of a legacy module via parallel subagents per file. Triggers on '/legacy-extract <module>' or when user asks to 'витягти legacy', 'extract legacy module', 'дослідити старий код через subagents'.

2026-05-16
legacy-plan
Softwareentwickler

Use when running Step 3 of legacy refactoring protocol — generates PLAN.md for module migration. Two phases — skeleton (3 sections only) and finalize (full plan from green tests). Triggers on '/legacy-plan <module> --phase=skeleton', '/legacy-plan <module> --phase=finalize', or when user asks to 'написати скелет плану', 'дописати план з тестів'.

2026-05-16
legacy-screencast
Softwareentwickler

Use when preparing screencast recording for one of the four protocol steps. Generates an asciinema or vhs script with explicit narration markers. Triggers on '/legacy-screencast <step>' or when user asks to 'записати скрінкаст для Step 2', 'screencast script', 'демо запис'.

2026-05-16
legacy-spike
Computersystemanalytiker

Use when starting Step 1 of legacy refactoring protocol — runs a 2-4 hour feasibility spike for one Bounded Context. Triggers on '/legacy-spike <module>' or when user asks for a feasibility check, BC scope decision, or 'чи злітає переписування цього модуля'.

2026-05-16
legacy-tests
Softwarequalitätssicherungsanalysten und -tester

Use when running Step 3.2 of legacy refactoring protocol — generates contract/characterization/architecture tests for the new module before any cutover. Triggers on '/legacy-tests <module>' or when user asks to 'згенерувати тести для legacy', 'characterization tests', 'fix point of reference'.

2026-05-16
pdf-form-filler
Softwareentwickler

Fill PDF forms programmatically when the user has a fillable PDF and a set of values to populate. Use when the user provides a PDF with form fields and asks to fill them, when the user mentions "AcroForm" or "fillable PDF" workflows, or when batch-filling multiple forms with the same template. Use even if the user does not mention pdfplumber or pikepdf by name.

2026-05-16
audit-api-endpoint
Softwarequalitätssicherungsanalysten und -tester

Use when reviewing one HTTP endpoint for production readiness — checks status, Content-Type, body shape, cache headers, and surfaces health-vs-ready confusion. Black-box runtime probe, works on any HTTP server regardless of language (Go, Node, Python, Rust, etc.). Trigger when the user names a single URL to audit, asks about /health or /ready quality, or wants a structured audit report for one endpoint.

2026-05-16