Copyright (c) 2026 JG Systems Consulting Ltd. - MIT License (tooling); pack content under each source's own licence (see NOTICE).
What it is / who it's for
A great body of knowledge (SEBoK, the NASA SE Handbook, a modelling spec) is something you
read once and then can't recall when you need it mid-task. A knowledge pack turns it into a
skill your coding agent loads on demand: ask /sebok systems of systems and the agent reads the
right chapter and answers from reconstructed reference notes, not a hazy recollection.
Don't know which pack to ask? The catalogue ships one orchestrator: /se <question> routes your
systems-engineering question to the right pack or packs, reads them, and answers with pack and
chapter citations. It carries no source content and answers only when explicitly invoked.
This is a curated catalogue of those packs for systems engineers, built to one rule that makes
it different from "point a tool at a PDF":
Only genuinely open sources go in. Public-domain (NASA, DoD) and open-licensed (SEBoK, OMG
specs, CC content): never paywalled or all-rights-reserved material. Eligibility is enforced,
not assumed (see docs/SOURCE-VETTING.md), and every pack is
redistributable by construction (see docs/LICENSING.md).
Packs follow the open Agent Skills SKILL.md
standard, so they load in Claude Code, GitHub Copilot CLI, and Amp without modification.
The catalogue is also listed in VoltAgent/awesome-agent-skills
(PR #1272), a curated collection
of 1,000+ community agent skills.
Prerequisites
- Python 3.9+ (for the installer and pack validator).
- A host that reads Agent Skills (Claude Code, GitHub Copilot CLI, or Amp).
- No MCP server, API key, or licence tier is needed at runtime. Packs are plain Markdown.
Install with your AI agent
Copy the block below into your coding agent (Claude Code, Cursor, etc.). It will read this repo
and install the catalogue for you.
You are installing jgs-se-knowledge-packs, an open catalogue (MIT tooling) of
systems-engineering knowledge-pack skills by JG Systems Consulting Ltd.
Repository: https://github.com/jgsystemsconsulting/jgs-se-knowledge-packs (version 1.24.1).
Do this in order:
1. Read README.md, docs/skill-usage.md, and CHANGELOG.md so you understand what you are
installing. There are NO external prerequisites: packs are plain Markdown skills.
2. Run `python install.py --dry-run` and show me the pack list and target
(~/.claude/skills/jgs-se-knowledge-packs/). If it looks right, run `python install.py`.
3. Verify: list the installed packs under the target and confirm the count matches SKILLS.md.
4. Tell me to restart Claude Code, then try `/sebok systems of systems` or ask
`/se what are the EARS requirement patterns?`.
5. Pack CONTENT is under each source's own licence (see NOTICE); keep attribution intact.
Install with the skills CLI
The catalogue is listed on skills.sh.
With the Vercel skills CLI you can pull packs without cloning:
npx skills add jgsystemsconsulting/jgs-se-knowledge-packs -l # list available packs
npx skills add jgsystemsconsulting/jgs-se-knowledge-packs --skill sebok # install one pack
npx skills add jgsystemsconsulting/jgs-se-knowledge-packs --all # install the catalogue
Install manually
git clone https://github.com/jgsystemsconsulting/jgs-se-knowledge-packs.git
cd jgs-se-knowledge-packs
python install.py --dry-run # preview packs + target
python install.py # installs to ~/.claude/skills/jgs-se-knowledge-packs/
Default target is the vendor-namespaced ~/.claude/skills/jgs-se-knowledge-packs/ (honours
$CLAUDE_CONFIG_DIR). For a flat install (each pack directly under ~/.claude/skills/), pass
--flat. Restart your agent, then invoke a pack by its slug.
Use with other agents
The packs are open Agent Skills; install.py installs them into
agents beyond Claude Code, natively where the agent reads the format, or via a format
transform where it doesn't.
python install.py --list-agents # show every supported agent + its target
python install.py --agent openclaw # OpenClaw / Copilot CLI, native (full pack)
python install.py --agent codex # Codex / Gemini / Cursor, transform (SKILL.md index)
python install.py --agent all # all user-global agents at once
| Agent | Format | Target |
|---|
| Claude Code (default) · OpenClaw · GitHub Copilot CLI | native folder | ~/.<agent>/skills/jgs-se-knowledge-packs/<slug>/ |
| OpenAI Codex CLI | prompt | ~/.codex/prompts/<slug>.md |
| Gemini CLI | TOML command | ~/.gemini/commands/jgs-se-knowledge-packs/<slug>.toml |
| Cursor | project rule | ./.cursor/rules/<slug>.mdc |
Native agents get the whole pack (chapters + progressive disclosure); transform agents get the
always-loaded SKILL.md index inlined into one file. Full detail and the per-agent limitations:
docs/other-agents.md.
Catalogue
| Pack | Source | Licence | Tier | Status |
|---|
sebok | Guide to the SE Body of Knowledge v2.14 | CC BY-NC-SA 3.0 | 🟡 2 | ✅ live (36 chapters) |
nasa-se-handbook | NASA Systems Engineering Handbook | Public domain (US gov) | 🟢 1 | ✅ live (34 chapters) |
nasa-se-expanded | Expanded Guidance for NASA SE (SP-2016-6105-SUPPL Vol 1) | Public domain (US gov) | 🟢 1 | ✅ live (6 chapters) |
nasa-npr-7123 | NASA NPR 7123.1D: SE Processes & Requirements | Public domain (US gov) | 🟢 1 | ✅ live (9 chapters) |
nasa-risk | NASA Risk Management Handbook | Public domain (US gov) | 🟢 1 | ✅ live (7 chapters) |
nasa-system-safety | NASA System Safety Handbook v2 | Public domain (US gov) | 🟢 1 | ✅ live (7 chapters) |
nasa-pra | NASA PRA Procedures Guide 2nd ed (NASA/SP-2011-3421) | Public domain (US gov) | 🟢 1 | ✅ live (8 chapters) |
nasa-hsi | NASA Human Systems Integration Guide | Public domain (US gov) | 🟢 1 | ✅ live (9 chapters) |
nist-ai-rmf | NIST AI Risk Management Framework 1.0 | Public domain (US gov) | 🟢 1 | ✅ live (8 chapters) |
nist-ssdf | NIST Secure Software Development Framework 1.1 | Public domain (US gov) | 🟢 1 | ✅ live (5 chapters) |
nist-csf | NIST Cybersecurity Framework 2.0 | Public domain (US gov) | 🟢 1 | ✅ live (8 chapters) |
nist-sse | NIST SP 800-160 Vol 1+2: Systems Security Engineering | Public domain (US gov) | 🟢 1 | ✅ live (8 chapters) |
nist-800-37 | NIST SP 800-37 Rev 2: Risk Management Framework | Public domain (US gov) | 🟢 1 |
doe-o-413-3 (formerly doe-413-3b; old slug kept as a catalog alias) tracks the current DOE O 413.3 series edition (O 413.3C, which cancels O 413.3B Chg 7).
Machine-readable index: SKILLS.md · catalog.json.
Signposts (citation-only, not packs, zero reproduced content): omg-signpost points to
the OMG modelling specs; se-standards-signpost maps the whole SE standards landscape
(ISO/IEC/IEEE, INCOSE, SAE/EIA, ECSS, NATO, CMMI, NIST SP 800-160…) with each standard's
owner, whether it can be packaged, and where to get it. automotive-signpost points to the
automotive standards landscape (ISO 26262, ASPICE, SAE J3016, UNECE R155/R156) and
functional-safety-signpost points to the functional safety landscape (IEC 61508, ISO 26262,
ISO/SAE 21434, UL 4600, SAE J3016, ASPICE); both are citation-only, and the open path for
both is nhtsa-vehicle. avionics-signpost points at the avionics development assurance standards (DO-178C, DO-254, ARP4754B, ARP4761A) with faa-8110-49 as the open path. rail-signpost points at EN 50126, EN 50128, and EN 50129, with CSM regulations and ORR guidance as the free paths and no in-repo pack. space-signpost points at ECSS documents with catalogue links, free paths are existing NASA packs, and several needs have no in-repo pack. maritime-signpost points at SOLAS, ISM, ISPS, COLREGS, and class rules; free paths are USCG material (including the Navigation Rules PDF and eCFR Title 46), UK MCA notices, and class portals; no in-repo pack covers these instruments. They exist because most SE standards
are paywalled or all-rights-reserved (see docs/SOURCE-VETTING.md).
Orchestrator (not a pack, zero source content): se is an explicit /se <question> entry
point that routes your systems-engineering question through a curated topic, agency, and
deliverable map to one to four packs, reads them with you, and answers with pack and chapter
citations. Use it when you do not know which /<slug> to ask.
Usage
/sebok # load core frameworks for reference
/sebok systems of systems # routes to the right chapter, answers from the pack
/sebok ch20 # load a specific chapter
/se what are the EARS requirement patterns? # orchestrator: routes to the right pack(s), answers with citations
/se how do NASA and DoD differ on risk acceptance? # broad question: one plan, the packs read together
Each pack ships an always-loaded SKILL.md index plus on-demand chapters/, a glossary,
patterns, and a cheatsheet. Full guidance: docs/skill-usage.md.
How a pack is built
Vet the source (tier + licence, not on the Excluded list) → extract text (built on the
MIT-licensed jgs-reference-skill
engine: a fork of book-to-skill) → map
structure to chapter offsets → generate reference-depth chapters in parallel + glossary /
patterns / cheatsheet + the SKILL.md index → attach provenance (PACK.yaml + a LICENSE
reproducing the source's terms) → validate. Contract: docs/PACK-SPEC.md.
Contributing: CONTRIBUTING.md.
Licensing
Two separable layers:
- Tooling & scaffolding: MIT (JG Systems Consulting Ltd.).
- Pack content: each pack carries its source's licence in
packs/<slug>/LICENSE,
independent of the repo MIT licence. A CC BY-NC-SA source yields a CC BY-NC-SA pack.
Per-pack source attributions are in NOTICE. Why every pack is lawful to reconstitute
and redistribute (including why no source-material links are published) is set out in
docs/LICENSING.md.
No purchase or licence key is needed to use this tooling. For commercial licensing
questions, see labs.jgsystemsconsulting.com/licensing.html.
Support & security
Version
See CHANGELOG.md. Current: 1.24.1.