Zarządzanie i produkt · Planowanie
pending
/pending
Pokazuje, jakie zadania z agendy są gotowe do wykonania, albo przekazuje ten przegląd innemu procesowi.
- Autor
- yuri-xyz ↗
- Repozytorium
- skills ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 03.11.2026
Ocena i bezpieczeństwo
100/ 100Klasa: dobry
Format25 / 25
Frontmatter z nazwą i opisem, poprawna nazwa, opis do 1024 znaków, nazwa zgodna z folderem.
Odkrywalność25 / 25
Opis mówi, kiedy użyć skilla, ma sensowną długość i jest w 3. osobie.
Treść35 / 35
Co najmniej 150 słów i do 500 linii, nagłówki, listy kroków, przykłady albo kod.
Higiena15 / 15
Bez zaślepek typu TODO, bez ścieżek Windows, z licencją.
Ryzykobrak sygnałów w skanie
- Pliki w paczce
- 2
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Przestarzałość
Brak sygnałów. Repozytorium zmienione 03.11.2026, bez odwołań do wycofanych modeli i API.
Licencja
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Pokazuje, jakie zadania z agendy są gotowe do wykonania, albo przekazuje ten przegląd innemu procesowi.
Oryginalny opis (EN)
Use when the user asks what pending agenda work is ready to do, or when another skill, such as `/autopilot`, needs the ready-work overview as a step.
Jak działa
- Objective
- Protocol
- Readiness
- Importance
- Independence
- Areas
- Output Format
Sekcje z nagłówków SKILL.md, w oryginale · 111 linii, 729 słów
Podgląd SKILL.md
--- name: pending description: Use when the user asks what pending agenda work is ready to do, or when another skill, such as `/autopilot`, needs the ready-work overview as a step. --- # Pending ## Objective Give the user a quick overview of the ready work waiting in the repository's `.agenda` workspace, so they can pick what to do next. This skill only reports; it does not create, edit, rename, or implement agenda entries. Follow `/agenda` for the workspace layout, filename statuses, and the not-ready disclaimer shape. ## Protocol 1. Find the repository root and read `.agenda/tasks` and `.agenda/plans`. If neither exists or both are empty, report that there is no agenda. 2. Collect pending entries: tasks with `todo` or `in_progress` status, and plans with `planned` or `active` status. Skip `done` entries. Read each pending entry's content, not just its filename. 3. Filter out every entry that is not ready (see Readiness). 4. Assign each remaining entry an importance and a feature or area, and mark whether it is independent. 5. Report them in the output format below. ## Readiness Classify each pending entry with `/agenda-readiness` and keep only the ones it rates **ready**. Leave out the rest, including entries an agent is already working on, manual entries that act on live external state and need your oversight, and exploratory or intentionally deferred entries that are being considered rather than given the go-ahead. Count the entries left out for the footer. Never list an entry that carries an `Assignee` claim (see `/agenda`), live or stale: this overview is for work that is ready and not yet started. Count claimed entries in the footer as left out, and say how many have a stale claim so abandoned work is not forgotten. An entry left unfinished with no claim and no agent working on it is listed like any other ready entry; note its stale worktree or session in its summary so it is resumed rather than restarted. ## Importance Judge how important each entry likely is from its content and the repository context; do not just copy any stated priority without checking it still holds. - 🔴 **High**: correctness, security, data-loss, or broken-workflow issues, work that unblocks other entries, and unfinished `active` or `in_progress` work left stale. - 🟡 **Medium**: meaningful improvements, features, or cleanups with clear value but no urgency. - 🟢 **Low**: polish, minor refactors, nice-to-haves, and speculative ideas. ## Independence Mark an entry 📦 **independent** when it can be done in parallel with the rest of the pending work without coordination beyond trivial steps: - it does not depend on another pending entry, and no other pending entry depends on it; - it is unlikely to edit the same files, modules, or shared configuration as another ready entry, so parallel work would not conflict; - it needs no shared migration, lockfile or input update, live system-state change, or approval that other work also waits on. Check this against the entries' content and the files they name, not only their areas. When unsure, leave the mark off. ## Areas Group entries by the feature, subsystem, or area they concern, such as networking, desktop, or build tooling. Derive area names from the entries and the repository's own structure, keep them short, and merge areas that would hold a single closely related entry into a broader one when that reads better. ## Output Format Order areas by their most important entry, then by entry count. Within each area, order entries by importance, highest first. Use exactly this shape: ```markdown 📋 **Pending work** **Area name** - 🔴 Task 3: Title — one-line summary of what it does and why it matters - 🟡📦 Plan 2: Title — one-line summary (active, item 2 of 5) **Other area** - 🟢📦 Task 7: Title — one-line summary 📦 independent: can run in parallel with other work ⏸️ 4 entries left out (claimed, in progress, manual, blocked, waiting, or not given the go-ahead); 1 claim is stale ``` - Label entries `Task <id>` or `Plan <id>` using their numeric ID. - Put 📦 directly after the importance emoji, with no space, for independent entries. Show the 📦 legend line only when at least one entry carries the mark. - For `in_progress` tasks and `active` plans, note progress in parentheses when it is visible from item statuses or checklists. - Omit the footer when nothing was filtered out, and the stale-claim clause when no claim is stale. If nothing ready remains, output exactly this instead, keeping the footer when blocked entries exist: ```markdown 🟣 **No pending work** ```
Pliki w paczce (2)
- SKILL.md 5 KB
- agents/openai.yaml 1 KB
Wersje
| Repozytorium | Wynik | Licencja |
|---|---|---|
| Ta strona · yuri-xyz/skills | 100 | MIT |
| M-reza-kd/lead-flow-plugin | 95 | MIT |
Migawka bazy z 13.11.2026 · wersja paczki 8fc23c399a4a (skrót SHA-256 całej paczki)
Pobrania w ostatnich 30 dniach: 0 · ranking
Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.
Podobne skille
- 100autopilot
Pracuje przez gotowy backlog .agenda: wybiera ustalone zadania kodowe, implementuje je i zatwierdza każdą jednostkę. - 100agenda-clean
Porządkuje backlog .agenda w repozytorium: usuwa mocno nieaktualne zadania i plany, a te, które są jeszcze ważne, aktualizuje. - 90run
Autonomicznie prowadzi projekt: wybiera gotowe zadania z listy oczekujących i wykonuje je, aż nie zostanie nic do zrobienia. - 100project-overview
Pokazuje przegląd wybranego projektu albo wszystkich projektów użytkownika i krótkie podsumowanie ich stanu. - 100file-todos
Prowadzi cykl życia plikowych TODO w todos/: jeden plik markdown na uwagę z przeglądu lub dług techniczny, od pending przez ready do complete, z triażem i równoległym rozwiązywaniem. - 100brainstorm
Rozmawia z użytkownikiem, żeby wskazać kolejne najbardziej produktywne jednostki pracy w projekcie, i zapisuje je w project.md w sekcji Pending Work.