LARi-UQAC/ResearchTools
AI-assisted toolbox for work auditing (LaTeX, Scopus validation, paper/thesis/report auditing), software design, CAD design, PCB design, local LLM with observer and tracker, built on Claude Code agents/skills/commands with mirrors to other
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit LARi-UQAC/ResearchTools/.claude ./rig-researchtools # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit LARi-UQAC/ResearchTools/.agents/skills/aider-setup .claude/skills/aider-setup $ npx degit LARi-UQAC/ResearchTools/.agents/skills/deliberation .claude/skills/deliberation $ npx degit LARi-UQAC/ResearchTools/.agents/skills/drawio2tikz .claude/skills/drawio2tikz $ npx degit LARi-UQAC/ResearchTools/.agents/skills/extract-contributions .claude/skills/extract-contributions $ npx degit LARi-UQAC/ResearchTools/.agents/skills/extract-futureworks .claude/skills/extract-futureworks $ npx degit LARi-UQAC/ResearchTools/.agents/skills/extract-paper-idea .claude/skills/extract-paper-idea $ npx degit LARi-UQAC/ResearchTools/.agents/skills/extract-statistic .claude/skills/extract-statistic $ npx degit LARi-UQAC/ResearchTools/.agents/skills/form-service .claude/skills/form-service $ npx degit LARi-UQAC/ResearchTools/.agents/skills/geolocalisation .claude/skills/geolocalisation $ npx degit LARi-UQAC/ResearchTools/.agents/skills/latex-hygiene .claude/skills/latex-hygiene $ npx degit LARi-UQAC/ResearchTools/.agents/skills/loop-engineer .claude/skills/loop-engineer $ npx degit LARi-UQAC/ResearchTools/.agents/skills/narrative-cv .claude/skills/narrative-cv $ npx degit LARi-UQAC/ResearchTools/.agents/skills/obsidian-cli .claude/skills/obsidian-cli $ npx degit LARi-UQAC/ResearchTools/.agents/skills/opt-local-stt-vram .claude/skills/opt-local-stt-vram $ npx degit LARi-UQAC/ResearchTools/.agents/skills/opt-local-vram-llm .claude/skills/opt-local-vram-llm $ npx degit LARi-UQAC/ResearchTools/.agents/skills/paper2talk .claude/skills/paper2talk $ npx degit LARi-UQAC/ResearchTools/.agents/skills/recommendation-letter .claude/skills/recommendation-letter $ npx degit LARi-UQAC/ResearchTools/.agents/skills/rt-observe .claude/skills/rt-observe $ npx degit LARi-UQAC/ResearchTools/.agents/skills/scholar-evaluation .claude/skills/scholar-evaluation $ npx degit LARi-UQAC/ResearchTools/.agents/skills/scientific-writing .claude/skills/scientific-writing $ npx degit LARi-UQAC/ResearchTools/.agents/skills/scopus .claude/skills/scopus $ npx degit LARi-UQAC/ResearchTools/.agents/skills/word2latex .claude/skills/word2latex $ npx degit LARi-UQAC/ResearchTools/.claude/skills/aider-setup/config/skills/test-driven-development .claude/skills/test-driven-development
$ curl -fsSL --create-dirs -o .claude/agents/abstract-writer.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/abstract-writer.md $ curl -fsSL --create-dirs -o .claude/agents/authoring-loop.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/authoring-loop.md $ curl -fsSL --create-dirs -o .claude/agents/bib-cleaner.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/bib-cleaner.md $ curl -fsSL --create-dirs -o .claude/agents/cover-paper.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/cover-paper.md $ curl -fsSL --create-dirs -o .claude/agents/latex-writer.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/latex-writer.md $ curl -fsSL --create-dirs -o .claude/agents/litreview-updater.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/litreview-updater.md $ curl -fsSL --create-dirs -o .claude/agents/local-coder.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/local-coder.md $ curl -fsSL --create-dirs -o .claude/agents/local-writer.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/local-writer.md $ curl -fsSL --create-dirs -o .claude/agents/narrative-cv-writer.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/narrative-cv-writer.md $ curl -fsSL --create-dirs -o .claude/agents/paper-auditor.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/paper-auditor.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer-response.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/reviewer-response.md $ curl -fsSL --create-dirs -o .claude/agents/scopus-auditor.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/scopus-auditor.md $ curl -fsSL --create-dirs -o .claude/agents/scopus-researcher.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/scopus-researcher.md $ curl -fsSL --create-dirs -o .claude/agents/submit-checker.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/submit-checker.md $ curl -fsSL --create-dirs -o .claude/agents/talk-builder.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/talk-builder.md $ curl -fsSL --create-dirs -o .claude/agents/thesis-auditor.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/thesis-auditor.md $ curl -fsSL --create-dirs -o .claude/agents/thesis-proposal-auditor.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/thesis-proposal-auditor.md $ curl -fsSL --create-dirs -o .claude/agents/thesis-to-paper.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/thesis-to-paper.md $ curl -fsSL --create-dirs -o .claude/agents/word-to-latex.md https://raw.githubusercontent.com/LARi-UQAC/ResearchTools/main/.claude/agents/word-to-latex.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} Skills (23)
aider-setupdeliberationdrawio2tikzextract-contributionsextract-futureworksextract-paper-ideaextract-statisticform-servicegeolocalisationlatex-hygieneloop-engineernarrative-cvobsidian-cliopt-local-stt-vramopt-local-vram-llmpaper2talkrecommendation-letterrt-observescholar-evaluationscientific-writingscopusword2latextest-driven-development
Subagents (19)
| abstract-writer | Use to draft or refresh a paper's abstract from its own content, or a UQAC thesis' Resume (French) + Abstract (English) pair, grounded in the paper's own extracted contribution/method/results/limitati |
| authoring-loop model: fable | ScholarEval-gated authoring improvement loop. Define a subject, author with the matching authoring agent (scopus-researcher / latex-writer / reviewer-response, etc.) on the best cloud model, audit the |
| bib-cleaner | Use when the user provides a `.bib` file and wants it validated, deduplicated, normalized, and enriched with missing DOIs. Produces a cleaned `.bib` file and a report. |
| cover-paper | Use when a paper is about to be submitted to a journal and needs its submission package: (a) a Cover Letter, standalone or hidden inside the main `.tex`, (b) for Elsevier / Springer / Wiley / MDPI onl |
| latex-writer | Use for LaTeX work: academic papers (IEEE/Springer/Elsevier), Beamer slides, TiKZ technical diagrams, thesis documents. |
| litreview-updater model: sonnet | Use to incrementally refresh an EXISTING literature review (from a prior /litreview run) with newly published papers, instead of re-running the full review. Searches Scopus + Consensus for papers sinc |
| local-coder model: haiku | Use for local code generation: implementing a function against a spec or failing test, refactor snippets, boilerplate, test scaffolds, and small deterministic scripts. Generates over the deterministic |
| local-writer model: haiku | Use for high-token repetitive writing (docstrings, inline code comments, Markdown documentation, CHANGELOG/README drafting) AND as the keeper of TWO memories: the Obsidian vault, where it writes every |
| narrative-cv-writer | Use to draft, refresh, or tailor the narrative 'CV descriptif' (FRQ CV-FRQ, or the structurally identical tri-agency CIHR/NSERC/SSHRC CV commun des trois organismes) to a specific grant competition, m |
| paper-auditor | Use when the user provides a complete scientific paper (`.tex`) and wants a full content audit: Introduction, Litterature review, Methodology, Results, Discussion, Future Works, and all references val |
| reviewer-response | Use when the user provides reviewer comment files (one per reviewer) and wants point-by-point LaTeX response letters generated, with corrections applied directly in the paper using `\\added{}`/`\\dele |
| scopus-auditor | Use when the user provides an existing review text (LaTeX, plain text, or pasted) and wants references validated, errors flagged, and an executable improvement plan produced. |
| scopus-researcher | Use for autonomous literature reviews: finding, validating, and summarizing academic papers from Scopus on a given topic. Domain scope set by the active profile in profiles/<name>.yaml (engineering de |
| submit-checker | Use when the user wants to check whether a paper is ready for submission to a specific journal. Produces a submission checklist with pass/fail status for all requirements. |
| talk-builder | Use to turn an accepted paper into a conference talk: a deck on the lab gabarit with timed speaker notes, in PowerPoint, LaTeX Beamer, or a self-contained web page. Triggered by the `paper2talk` skill |
| thesis-auditor | Use when the user provides a UQAC Master's or PhD thesis (LaTeX, `uqac.cls`) and wants a full institutional and academic audit: front matter compliance, hypothesis flow, chapter structure, references, |
| thesis-proposal-auditor | Use when the user provides a UQAC Master's or PhD **thesis proposal** (LaTeX, `uqac.cls`) and wants a full institutional and academic audit of the proposal (not the final thesis): short Introduction ( |
| thesis-to-paper | Use to integrate a student thesis (and its derived conference articles) into one journal manuscript in LaTeX — typically an invited extension of an accepted conference paper (special issue, e.g. NEWCA |
| word-to-latex | Use to convert a Word `.docx` template (Mitacs, CRSNG, FRQNT, UQAC, partner forms) into a faithful LaTeX source. Triggered by the `/word2latex` skill and by any task that involves matching a generated |
Slash commands (29)
/abstract/auditpaper/auditreview/auditthesis/bibclean/concis/ctx/cv/doc/fetchform/focus/geolocalisation/latex/litreview/litupdate/loopdev/opt-local-stt-vram/opt-local-vram-llm/recommendation-letter/ref/replyreviewer/rt-dashboard/slim/submitcheck/talk/test/texcheck/tikz/word2latex
Plugins (5)
code-review@claude-plugins-officialpr-review-toolkit@claude-plugins-officialgithub@claude-plugins-officialai-firstify@techwolf-ai-firstengineering@knowledge-work-plugins
Similar rigs
caitengwei/my-skills
我的 skills 聚合仓库
Skill Collector 3.0k tok ·
K-Dense-AI/claude-scientific-writer
A general purpose scientific writer
Skill Collector 8.2k tok ·
spacering-net/codeg
Collaborative multi-agent AI coding workspace: aggregate sessions from Claude Code, Codex, OpenCode, Pi, Grok Build, etc. Desktop app, self-hosted server, or Docker.
Skill Collector 2.8k tok ·
brunob54/superpowers-orchestrator
Claude Code plugin that runs a development pipeline autonomously: spec → plan → plan-review rounds → batched subagent implementation → code-review rounds, stopping only on errors. Fork of obra/superpowers.
Skill Collector 638 tok ·