report
/report
Tworzy raport do udostępnienia, PDF lub HTML z prezentacją, sekcję manuskryptu albo opracowanie z gotowych analiz Mycelium.
- Autor
- arjunrajlaboratory ↗
- Repozytorium
- mycelium ↗
- Licencja
- MIT
- Ostatnia zmiana repozytorium
- 17.10.2026
Ocena i bezpieczeństwo
- Pliki w paczce
- 2
- Skrypty
- 0
- Komentarze HTML
- 0
- Adresy URL w SKILL.md
- 0
Brak sygnałów. Repozytorium zmienione 17.10.2026, bez odwołań do wycofanych modeli i API.
MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.
Co robi i kiedy
Tworzy raport do udostępnienia, PDF lub HTML z prezentacją, sekcję manuskryptu albo opracowanie z gotowych analiz Mycelium.
Oryginalny opis (EN)
Create a shareable report — a PDF, or a self-contained HTML report with a companion slide deck — a manuscript section, or a structured writeup from completed Mycelium analyses. Load the project's report templates, provenance, section guidance, and QC checklist. Use when the user wants a document artifact for collaborators or publication. Do not use for conversational explanation, standalone figures, performing analysis, ingestion, setup, or brainstorming.
Jak działa
- Routing
- Post-action
- What this skill is NOT for
Podgląd SKILL.md
--- name: report description: > Create a shareable report — a PDF, or a self-contained HTML report with a companion slide deck — a manuscript section, or a structured writeup from completed Mycelium analyses. Load the project's report templates, provenance, section guidance, and QC checklist. Use when the user wants a document artifact for collaborators or publication. Do not use for conversational explanation, standalone figures, performing analysis, ingestion, setup, or brainstorming. --- # Mycelium — Report Resolve bundled `skills/` and `network/` paths relative to the Mycelium plugin root—the ancestor containing `.codex-plugin/` and `.claude-plugin/`. Resolve reports and `.living/` paths relative to the user's repository. Generate a structured report from an analysis, routing to the appropriate report convention pack installed in this repository. This command is a thin router. The substantive workflow — the planning brief, the memory cheatsheet, the manifest, the draft, the three blind sub-agent reviewers, and the recompile gate — lives in the convention pack's own `analysis-conventions.md`. The router's job is to pick the right convention pack and hand off; do not duplicate or summarize the convention's phases here. ## Routing 1. **Read `.living/conventions/ACTIVE_CONVENTIONS.yaml`** to see what is installed. 2. **If `report-generator` is installed** (check `.living/conventions/report-generator/`): first check for staleness — compare the `version:` field of the installed `.living/conventions/report-generator/CONVENTION_PACK.yaml` against the bundled `network/conventions/report-generator/CONVENTION_PACK.yaml` (resolved relative to the plugin root). Plugin updates refresh only the bundled `network/` tree, never the copy installed in `.living/conventions/`, so an upgraded plugin routinely carries a newer pack than the repository is running. If the bundled version is newer, tell the user which versions differ and offer to refresh before drafting — `python skills/core/scripts/install_convention.py --name report-generator` replaces the installed copy in place (ask rather than overwrite silently: installed packs may carry local edits). Next, check the report linter once per report: `python skills/core/scripts/check_linter_versions.py "scitexlintr>=0.2"` reports the installed version and the latest on PyPI and never installs. If it exits 1 (missing or too old), install with the command it prints. If it reports an update, tell the user and upgrade only with their agreement, before drafting rather than mid-report, because a newer linter can add rules. Then follow the installed pack's `analysis-conventions.md` as the entry point. That file orchestrates the full phase-based flow (planning brief → memory consultation → section outline → manifest → draft → worked-example gate → three blind sub-agent reviewers → recompile + log → optional headline preview). The planning brief also picks the output format — LaTeX PDF (default) or one self-contained HTML file holding the report and a 10–20 slide companion deck — and the pack's `html-conventions.md` covers the HTML format end to end (template, value spans, figure and data inlining, the HTML gate, and the Phase 9 slides). The convention pack carries the three TeX template variants (`overview`, `comprehensive`, `overview-supplement`) and the HTML template with a worked example, the section-by-section craft notes (`references/section-guide.md`), the sub-agent prompts (`references/phase-prompts.md`), and the provenance/style QC checklist (`qc-checklist.md`). 3. **If other report conventions are installed** (future packs like `internal-memo`, etc.): present the user with the available report styles and let them choose. Each pack carries its own `analysis-conventions.md` describing its flow. 4. **If no report convention is installed**: fall back to core `skills/core/references/writing-conventions.md` and `skills/core/templates/report-template.tex`. The fallback structure is a single-document Title / Abstract / Introduction / Methods (Data / Analysis / Statistical Methods) / Results / Discussion / References — no phase-based flow, no sub-agent reviewers. This path exists so the skill is usable in a project that has not yet installed the convention pack; suggest installing `report-generator` if the user expects to write more than one report. ## Post-action After the chosen convention pack's flow completes (or after a fallback draft is finalized), update the analysis manifest entry with the report status and the path to the PDF or HTML file. Log any decisions or learnings from the writing process to `.living/decisions.md` and `.living/learnings.md` per the core post-action protocol — particularly any new failure modes the sub-agent reviewers surfaced, so the next report can probe for them upfront. For each learning: Derive `mitigation_type` by the rule in `skills/core/templates/learning-entry.md`: `structural` if a shipped test, assertion, type constraint, frozenset, or schema validation enforces it; otherwise `convention` if a `.living/conventions.md` entry covers it; otherwise `ambient-awareness`. ## What this skill is NOT for - Running the analysis itself — that is `/mycelium:analyze`. - Reviewing existing code or analysis decisions — that is `/mycelium:review`. - Open-ended idea generation — that is `/mycelium:ideas`. - Repo initialisation — that is `/mycelium:core init`. - Quick chat-based summaries — answer those conversationally without invoking this skill.
Pliki w paczce (2)
- SKILL.md 5 KB
- agents/openai.yaml 1 KB
Wersje
| Repozytorium | Wynik | Licencja |
|---|---|---|
| Ta strona · arjunrajlaboratory/mycelium | 100 | MIT |
| petrbouda/jeffrey | 100 | Apache-2.0 |
| petrbouda/jeffrey | 100 | Apache-2.0 |
| onmax/comark-ui | 100 | MIT |
| renzrollon/interlock | 100 | MIT |
| digital-stoic-org/agent-skills | 100 | MIT |
| 90p-labs/harness | 100 | MIT |
| emaballarin/ccplugins | 100 | MIT |
| quantskills/agent-quantspace | 100 | GPL-3.0 |
Migawka bazy z 13.11.2026 · wersja paczki 5c2630b3b892 (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
- 100sci-html
Zamienia artykuł PDF, notatki lub Markdown w interaktywną prezentację HTML z rysunkami i stronami sekcji. - 100manuscript-review
Przeprowadza audyt manuskryptu przed publikacją, tworzy raport z refaktoryzacji sekcji i sprawdza higienę cytowań oraz gotowość pracy do wysłania. - 100manuscript-setup
Tworzy projekt nowego manuskryptu akademickiego w LaTeX lub Word z osobnym plikiem na każdą sekcję, folderami bibliografii i rysunków oraz plikiem konfiguracji. - 100v1-html-it
Tworzy samodzielną stronę HTML z raportem, pulpitem lub interaktywnym wyjaśnieniem, gotową do udostępnienia. - 95build-report
Zamienia analizę procesu w dopracowany raport dla klienta: najpierw deck HTML do zatwierdzenia, potem zgodny z nim PDF. - 95results-report
Pisze zwięzły raport z wynikami badania behawioralnego dla promotora, zespołu lub własnych zapisów, w Markdown lub HTML.