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.
/plugin marketplace add superduke/ganvilnpx agentmods add plugins/superduke/ganvil/marketplacegit clone --depth 1 https://github.com/superduke/ganvilGrade A, and why
ganvil 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 2d 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": "ganvil",
"owner": {
"name": "superduke"
},
"plugins": [
{
"name": "ganvil",
"source": "./ganvil",
"description": "Long-running autonomous coding harness with Planner-Generator-Evaluator architecture and Sprint decomposition.",
"version": "1.3.0",
"author": {
"name": "superduke"
}
}
]
}
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.
- 2d ago First seen · 18 lines scan A 8ff87ad8dafb
ganvil is a plugin published in the GitHub repository superduke/ganvil (3 stars, last pushed 1mo ago), 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 2026-08-31.
Other plugins, from other repositories
festival
Festival Methodology: goal-oriented project management for human-AI development workflows. Provides fest and camp CLI integration, automatic installation, and methodology guidance.
visionclaw-agent
VisionClaw Agent — multi-tenant agentic AI platform exposing 243 production tools, 16 specialist personas, hybrid-search Skill-RAG, and a self-improving skill registry over MCP. One-click access to project management, Drive/email/calendar, web research, code execution, multi-channel messaging, schema-aware databases…
iterative-development
An iterative implementation methodology that pairs with superpowers. Extracts requirements with proof obligations from large spec collateral, defines a walking skeleton that closes a real journey, then loops through audited sprints building a behavior evidence corpus until an auditor confirms the product matches the…
spine
Understand any codebase, then ship changes to it. Read-only comprehension tools (maprepo, blastradius, explainsymbol, investigate, localize, regressiongaps, rootcause, docsfor) hand you engineering decisions — what breaks if you change X, what's untested, where a ticket or bug lands, which docs describe it — from.
spine
Spine's Claude Code plugin marketplace — delegate a ticket, get a reviewed PR.
dev-flow
An end-to-end product-development skill suite: one filesystem contract (.workflow/meta.json) and 43 skills that take an idea to production — web (Next.js 16), mobile (Expo/RN), an eve agent engine, Linear/scrum, plus GDPR/AI-Act and Vercel-cost pre-deploy gates.