hashgraph-online/awesome-codex-plugins / plugins/alpoxdev/hypercore/skills/hyper-research
hyper-research
Verified[Hyper] Use this skill when the user asks for multi-source, cited research - fact-finding, comparisons, market or trend analysis, or evidence-backed recommendations across live web, official docs, GitHub, and local repo sources - saved as a markdown report. Do not use for one-source lookups, rewriting, or implementation.
| File | Kind | Size |
|---|---|---|
| SKILL.ko.md | markdown | 20.6 KB |
| SKILL.md | skill md | 17.6 KB |
| assets/evals/hyper-research-cases.jsonl | other | 23.4 KB |
| references/channel-selection.ko.md | markdown | 5.7 KB |
| references/channel-selection.md | markdown | 5.2 KB |
| references/journal-format.ko.md | markdown | 7.6 KB |
| references/journal-format.md | markdown | 7.0 KB |
| references/latex-report.ko.md | markdown | 7.0 KB |
| references/latex-report.md | markdown | 6.4 KB |
| references/report-template.ko.md | markdown | 5.3 KB |
| references/report-template.md | markdown | 5.0 KB |
| rules/claim-verification.ko.md | markdown | 9.8 KB |
| rules/claim-verification.md | markdown | 8.5 KB |
| rules/expansion-and-convergence.ko.md | markdown | 6.7 KB |
| rules/expansion-and-convergence.md | markdown | 5.4 KB |
| rules/parallel-research.ko.md | markdown | 6.2 KB |
| rules/parallel-research.md | markdown | 5.3 KB |
| rules/research-workflow.ko.md | markdown | 8.3 KB |
| rules/research-workflow.md | markdown | 7.2 KB |
| rules/validation.ko.md | markdown | 11.6 KB |
| rules/validation.md | markdown | 10.1 KB |
| scripts/report-state.shexecutable | script | 16.5 KB |
| scripts/validate-report.shexecutable | script | 16.2 KB |
references/latex-report.md
View on GitHub# LaTeX Report
Use this reference only when the user asks for LaTeX or a typeset PDF. The Markdown report stays the deliverable of record; the `.tex` file is built from the finished report, never the other way around.
## Contents
- Build output
- Citation and ledger
- Compile order
- Log-first repair
- Korean typesetting
- Page review
- Missing toolchain
## Build Output
Build the deliverable beside the saved report in `.hyper/research/` as `[NN].slug.tex`, plus the compiled `[NN].slug.pdf` when the toolchain is present. The `.tex` source is a rendering of the finished Markdown report: same claims, same citations, same numbers. Any prose polish applied while adapting the report to LaTeX must not move a `\cite` and must not strengthen an uncertain claim. A claim that is hedged in the report stays hedged in the `.tex`; a claim that sits in the unresolved annex is not promoted into the findings to fill a page.
## Citation And Ledger
Every `\cite{Sn}` in the `.tex` file must resolve to a `[Sn]` row in the run's sources ledger (the journal's `sources.md`). Before compiling, list every cite key in the `.tex` file and check it against the ledger rows:
- A cite key with a ledger row is safe to compile.
- A cite key with no ledger row blocks delivery. Add the missing row from the opened source, or remove the cite and its claim, then recheck. Do not compile a report that cites a source the run never opened.
- Citation shape follows the report itself: `\cite{S1}` cites one row, `\cite{S1,S4}` cites two rows. Numbers and other figures still carry their `MEASURED`, `ASSUMED`, or `DERIVED` lineage from the report.
The `.tex` file also carries the bibliography those cite keys resolve to, one entry per cited ledger row: either a `thebibliography` item whose label is `S<n>`, or a `.bib` entry whose key is `S<n>`. Fill each entry from the ledger row's title, publisher, URL or path, and publication date, and leave out a field the ledger does not hold instead of inventing one. A cite key that resolves to a ledger row but has no bibliography entry still renders as `?` in the PDF, so add the entry in the same pass that adds the cite.
## Compile Order
Compile with `latexmk -xelatex` first. It drives the engine, the bibliography tool when the file uses a `.bib`, and the extra engine passes for references and cross-references by itself:
```sh
latexmk -xelatex [NN].slug.tex
```
When `latexmk` is not installed, run the engine, the bibliography tool, and the engine twice by hand:
```sh
xelatex [NN].slug.tex
biber [NN].slug
xelatex [NN].slug.tex
xelatex [NN].slug.tex
```
Substitute `bibtex` for `biber` when the bibliography is BibTeX-shaped. Skip the bibliography tool entirely when the file uses a `thebibliography` list: there is no `.bib` to process, and running `biber` then fails with `Cannot find 'control' file`. Two engine passes are required so that citation numbers and cross-references settle; with a bibliography tool they come after it, and with `thebibliography` they follow the first pass. A single pass leaves `?` markers in the PDF.
## Log-First Repair
Read the compiler log (`.log`) before touching the source, and fix the first `!` error first. TeX stops reporting once error recovery replaces a missing piece, so later errors are often symptoms of an earlier one. The routine loop is:
1. Run the compile order above.
2. Search the log for the first line starting with `!`. That line, and the lines around it, name the failure.
3. Fix only the cause of that first error, then recompile.
4. Repeat until the log shows no `!` line.
Never rewrite the whole document to silence a log: one error, one fix, one recompilation. A missing citation key is not a log error, so it is checked separately under Citation And Ledger.
## Korean Typesetting
Use the XeTeX engine so Korean text renders: `latexmk -xelatex` and `xelatex` already do. The `.tex` file must set a Korean font the engine can find (for example `fontspec` with a Korean font available on the machine) and must use a CJK-capable class or package. Three rules before delivery:
- The Korean font really is installed on the machine that compiles the report. Name it only if it is present; if no Korean font is installed, say so in the report instead of silently falling back to blank glyphs.
- Compile once and check the log: missing-glyph warnings (`Missing character`) mean the font or the class is wrong. Fix the font choice, not the text.
- English titles, commands, and URLs stay in the font that carries them; only the Korean prose needs the Korean font.
## Page Review
Look at the rendered pages, not only the compiler exit code. When `pdftoppm` is available, render the PDF to page images and read the ones that carry the parts most likely to break:
```sh
pdftoppm -png -r 100 [NN].slug.pdf page
```
Review at least the first page, one page with a comparison table, and one page with long Korean prose. Check that tables fit the text width, no glyph is blank, and citation numbers render instead of `?`. When `pdftoppm` is not available, review the `.tex` source against the Markdown report line by line and say in the report that page images were not reviewed.
## Missing Toolchain
When no TeX toolchain is installed (`latexmk` and `xelatex` both absent), deliver the `.tex` source and state plainly that the PDF was not built. The report must name the missing tool and the command that would build the PDF. Never claim a PDF exists when the compile never ran, and never attach a PDF path that has no file behind it.
## Sources
> Checked 2026-11-09.
| Claim | Source |
|---|---|
| `latexmk -xelatex` first with an engine, bibliography tool, and two-pass fallback, log-first repair of the first `!` error, page review with `pdftoppm`, every `\cite` mapped to a ledger row, and prose polish that neither moves a `\cite` nor strengthens an uncertain claim | `dbc-hbin/omp-ulw` at commit `74df40f` (2026-11-09); the source repository has no license, so the mechanism was re-authored in this package's own words and no text or code was copied |
| The saved Markdown report at `.hyper/research/[NN].slug.md` and its `[S<n>]` citations resolving to the journal's `sources.md` rows | `references/report-template.md` and `references/journal-format.md` in this package |
### Evidence grade
`LOCAL` - the compile route and the citation-to-ledger check are this package's own decision. The Markdown report remains the deliverable of record; the LaTeX route is an optional second rendering.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |