Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/FlyFission/nuclear-grade-context-engineeringWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/plugins/flyfission/nuclear-grade-context-engineering/nuclear-grade-context-engineering)<a href="https://agentmods.dev/plugins/flyfission/nuclear-grade-context-engineering/nuclear-grade-context-engineering"><img src="https://agentmods.dev/badge/plugins/flyfission/nuclear-grade-context-engineering/nuclear-grade-context-engineering.svg" alt="Measured on agentmods" height="20"></a>Grade A, and why
nuclear-grade scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
{
"name": "nuclear-grade",
"version": "0.6.0",
"description": "Risk-graded context engineering for AI-assisted software work — a catalog of agent skills, paste-ready command prompts, and change-record templates that keep agent work questioned, controlled, and reviewable.",
"author": {
"name": "FlyFission"
},
"homepage": "https://github.com/FlyFission/nuclear-grade-context-engineering",
"repository": "https://github.com/FlyFission/nuclear-grade-context-engineering",
"license": "MIT",
"keywords": [
"context-engineering",
"ai-agents",
"configuration-management",
"evidence",
"governance"
]
}
What it installs
The manifest is a name and a version. 29 skills, 28 commands, 5 agents travel with it, and installing the plugin installs all of them — 2,677 tokens a session between them. Each is measured on its own page, and each can be installed alone.
- Skill breaking-down-the-work A 84 tokens
- Skill deciding-who-decides A 80 tokens
- Skill organizing-project-folders A 94 tokens
- Skill rating-change-risk A 103 tokens
- Skill staying-on-mission A 90 tokens
- Skill briefing-an-agent A 99 tokens
- Skill checking-release-readiness A 67 tokens
- Skill closing-stale-packets A 84 tokens
- Skill learning-from-experience A 65 tokens
- Skill proving-claims A 65 tokens
- Skill questioning-attitude A 88 tokens
- Skill recording-what-an-agent-did A 90 tokens
- Skill reviewing-code-quality A 73 tokens
- Skill creating-change-records A 70 tokens
- Skill declaring-intent A 82 tokens
- Skill double-checking-before-acting A 78 tokens
- Skill handing-off-work A 65 tokens
- Skill reporting-shared-defects A 93 tokens
- Skill responding-to-incidents A 80 tokens
- Skill checking-legal-and-safety-wording A 81 tokens
- Skill checking-source-claims A 65 tokens
- Skill checking-what-a-change-affects A 72 tokens
- Skill choosing-what-to-control A 66 tokens
- Skill recording-a-known-good-version A 87 tokens
- Skill verifying-final-artifacts A 98 tokens
- Skill stress-testing-agent-changes A 103 tokens
- Skill vetting-outside-code-and-models A 85 tokens
- Skill tracking-deficiencies A 80 tokens
- Skill using-nuclear-grade A 79 tokens
- Command ng-breakdown A 0 tokens
- Command ng-close-packet A 0 tokens
- Command ng-decide-authority A 0 tokens
- Command ng-folders A 0 tokens
- Command ng-verify-artifact A 0 tokens
- Command ng-classify A 0 tokens
- Command ng-code-review A 0 tokens
- Command ng-context-pack A 0 tokens
- Command ng-drift-check A 0 tokens
- Command ng-prove A 0 tokens
- Command ng-question A 0 tokens
- Command ng-red-team A 0 tokens
- Command ng-report-defect A 0 tokens
- Command ng-trace A 0 tokens
- Command ng-trust-check A 0 tokens
- Command ng-deficiency A 0 tokens
- Command ng-incident A 0 tokens
- Command ng-intent A 0 tokens
- Command ng-learn A 0 tokens
- Command ng-legal-check A 0 tokens
- Command ng-new A 0 tokens
- Command ng-self-check A 0 tokens
- Command ng-ship-review A 0 tokens
- Command ng-source-check A 0 tokens
- Command ng-turnover A 0 tokens
- Command ng-baseline A 0 tokens
- Command ng-impact A 0 tokens
- Command ng-what-to-control A 0 tokens
- Agent judge A 63 tokens
- Agent observer A 62 tokens
- Agent planner A 58 tokens
- Agent educator A 62 tokens
- Agent runner A 66 tokens
What ships with it
1 file beside plugin.json in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 19 lines scan A f5a960e94ac8
nuclear-grade is a plugin published in the GitHub repository FlyFission/nuclear-grade-context-engineering (33 stars, last pushed yesterday), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other plugins, from other repositories
thumbgate
One 👎 becomes a hard rule the agent cannot bypass. Captures thumbs-down feedback, distills it into PreToolUse Pre-Action Checks, enforced across every future Claude Code session.
paceflow
PACE workflow enforcement system — Plan-Artifact-Check-Execute-Verify-Review protocol with hooks.
octomux
Octomux vendored skills and agent definitions.
yarbrain
A reviewable Markdown second brain for Codex and Claude Code.
agent-docs
Project docs that answer locally before the web — Agent Skills for AI coding agents.
nextjs
Official Next.js skills: adopt and optimize Cache Components, adopt Partial Prefetching, and verify runtime behavior against a running dev server.