Agent Skills by moonlight-lupin
agent-skills
A library of skills and plugins for Hermes Agent covering research, media, scraping, documents, DevOps and agent upkeep.
- 123 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- The skills run scripts, move and delete files and call external services with your keys.
- Dossier and enrichment skills handle personal data, so check that the sources are lawful.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/moonlight-lupin-agent-skillsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- The skills run scripts, move and delete files and call external services with your keys.
- Dossier and enrichment skills handle personal data, so check that the sources are lawful.
This is third-party code. Review the repository files before installing.
What it does
Skill folders are grouped by domain. Research: `deep-research` with iterative search and citations, `fact-checker`, `source-tracker`, `news-monitoring`. Creative: `image-studio` and `clips-studio` on fal.ai. File work: `fill-template`, `receipt-compiler`, `pdf2epub`, `marp-deck`. Agent upkeep: `log-analyzer`, `input-token-overheads`, `claude-plugin-converter`. Every skill is self-contained, so any one installs on its own. A separate `plugins/` section holds Hermes plugins: `skill-retrieval` injects only the six best-matching skills per turn using BM25, `lumen` maintains a wiki from conversation history and `dev-pipeline` sets a five-stage development process.
Who it is for. Hermes Agent users who want a ready set of working skills, and owners of other agents willing to adapt them to their own skill loader.
Good fit when
- An agent prepares a topic report with sources and claim verification.
- A batch of receipt photos must become a PDF for an expense claim.
- A skill library grew to a hundred entries and the prompt bloated with the list.
- Agent logs need analysis for error clusters, rate limits and timeouts.
Not a fit when
- The agent is not Hermes and tool names would need mapping, since the skills target its toolset.
- One narrow skill with deep documentation is needed, while this collection is broad and each folder short.
- Inputs must stay off third-party services, because some skills call fal.ai, Pexels and PDL.
Example request
Prepare a research report on this topic with sources, verify the main claims and save the links to the tracker.Limitations
The skills are written for Hermes tools (`web_search`, `web_extract`, `terminal`, `delegate_task`, `cronjob`). For Claude Code, Codex and other agents, tool names and loader conventions have to be mapped. Some skills need third-party keys such as fal.ai, Pexels and PDL, plus Python 3.11 or newer. `disk-cleanup` and the file organizer move and delete files, so run them only after reviewing the plan.
How to disable. Delete the relevant folders from `.agents/skills/` and `.claude/skills/`, and remove Hermes plugins from `~/.hermes/plugins/`, then restart the gateway.
FAQ
Do the skills work outside Hermes Agent?
Yes, with adaptation. The author says workflows and scripts are agent-agnostic, but tool names and loader conventions need mapping.
How do I install just one skill?
Add `--skill <name>` to `npx skills add`. Or use `gh skill install moonlight-lupin/agent-skills creative/image-studio/SKILL.md`.
What does the skill-retrieval plugin do?
It picks the six best-matching skills per turn by BM25 and drops the full list from the prompt. The number six is configurable in `plugin.yaml`.
How are the Hermes plugins installed?
With `hermes plugins install skill-retrieval`, or by copying a folder from `plugins/` into `~/.hermes/plugins/` and restarting the gateway.
Which skills cover research?
`deep-research` for cited reports, `fact-checker` for claim checks, `source-tracker` for a source database, `news-monitoring` for recurring digests and `notebooklm-mode` for answers grounded in your own sources.
Discussion
No questions or reviews yet.
Related
SkillsA pack of practical skills for Hermes Agent: notes, trips, code security, browser automation and interface themes.
Custom skills bundle for Hermes Agent
hermes-custom-skills
A monorepo of five skills for Hermes Agent: agent security hardening, skill auto-updates, skill search, cross-border expansion and Tencent Yuanbao chat
Thirteen stable and six experimental skills for Hermes Agent: stack diagnosis, migration, token and skill audits
Portuguese skills and guides for Hermes Agent: text editing, VPS operations, kanban, RAG and injection defense
A catalog of nearly two hundred skills: code delivery, review, debugging, AI governance, product work and agent coordination.
Official Hermes Agent plugins: Home Assistant, Spotify, Blender, Snyk, Langfuse, sandboxes and importers for other agents
README in short
The README holds a skill table with dependencies, three install routes and descriptions of six plugins. All skills validate against the Agent Skills specification with `npx skills-ref validate`.
README badge
Original SKILL.md
---
name: deep-research
description: "Use when the user asks for \"deep research\", a \"research report on\", a \"comprehensive analysis of\", to \"look into X in depth\", to \"write a report on X\", or any question needing multi-source synthesis beyond a single search — an autonomous multi-step deep research engine implementing an iterative Think → Search → Extract → Synthesize → Stop loop. The LLM drives every decision: what to search, what's relevant, what's missing, and when to stop. Produces a cited, magazine-quality report with inline citations, category-specific formatting, and research stats. For entity vetting/dossiers use entity-research; for news digests use news-monitoring; for source-grounded Q&A use notebooklm-mode."
license: MIT
metadata:
version: 1.9.2
author: moonlight-lupin
platforms: [linux, macos, windows]
hermes:
tags: [research, deep-research, report, synthesis, iterative, citations, evidence-basis, provenance]
related_skills: [news-monitoring, entity-research, notebooklm-mode, youtube-topic-research, fact-checker, source-tracker]
---
# Deep Research — Iterative Research Engine
An autonomous, multi-step research engine that performs exhaustive information
gathering and synthesis. Unlike a single `web_search`, this skill implements an
iterative loop where the agent plans, searches, extracts, synthesizes, and
decides when to stop — producing a cited report with structured evidence,
source quality tiers, and explicit gaps/contradictions sections.
See `references/structured-evidence-format.md` for the evidence.json layer and
`references/docsgpt-concepts.md` for adaptive-depth concept mapping.
## When to use
- User asks for "deep research", "research report", "comprehensive analysis"
- User wants a written report on a topic (not just a quick answer)
- Question requires multi-source synthesis with citations
- User says "look into X in depth" or "write a report on X"
## When NOT to use
- **Entity vetting/dossiers** → use `entity-research` (has sanctions screening, structured lenses)
- **Recurring news digests** → use `news-monitoring` (has cron, dedup, multi-language)
- **Source-grounded Q&A from collected sources** → use `notebooklm-mode` (has vault + RAG)
- **Quick factual question** → just use `web_search` directly
- **Single-source extraction** → use `web_search` + `web_extract`
## Architecture