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.
npx skills add Encod3d-Sec/TORCH --skill wiki-arsenalgit clone --depth 1 https://github.com/Encod3d-Sec/TORCHWrote 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/skills/encod3d-sec/torch/wiki-arsenal)<a href="https://agentmods.dev/skills/encod3d-sec/torch/wiki-arsenal"><img src="https://agentmods.dev/badge/skills/encod3d-sec/torch/wiki-arsenal/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/encod3d-sec/torch/wiki-arsenal"><img src="https://agentmods.dev/badge/skills/encod3d-sec/torch/wiki-arsenal.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00187 | $0.01069 |
| Opus 5 | $0.00093 | $0.00535 |
| Sonnet 5 | $0.00037 | $0.00214 |
| Haiku 4.5 | $0.00019 | $0.00107 |
Grade A, and why
wiki-arsenal 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 10d 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.
How it starts
The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
wiki-arsenal
The fast, wiki-first lookup engine for "what do I use against this surface". Runs the four
knowledge areas in parallel so a deep lookup is one wall-clock, not four serial reads. arsenal
delegates here; the hunt-* skills each carry their own wiki-first qmd_query (MCP-independent) and
can hand off here for a fast parallel lookup. Never hand-roll from memory when the wiki has the answer.
Input: a surface, service, or vuln-class (e.g. Jenkins on 8080, SSRF, Kerberoasting).
0. Cache check first (0 tokens on a repeat)
Slug the surface (lowercase, non-alnum -> -). If targets/<active-eng>/arsenal/<slug>.md
exists, read and return it. Do not re-spend. (<active-eng> = the dir named in targets/active.md.)
Mode: quick (DEFAULT - fire it constantly)
One mcp__wiki-search__qmd_query over the whole index (add a qmd_search keyword pass when the
surface is an exact product/CVE string). Group the hits under the four areas and return each as
path -> one-line snippet:
- Techniques (
wiki/techniques/) - Payloads (
wiki/payloads/) - Tools (
wiki/tools/) - Cheatsheets (
wiki/cheatsheets/)
Cost ~1-2k tokens, no subagents. This is what arsenal calls by default and what you fire on
every new surface to raise wiki coverage cheaply. Stop here unless the surface is worth deep prep.
Mode: deep (opt-in - "deep"/"full arsenal", or a whole service/target worth prepping)
Dispatch FOUR parallel subagents in a SINGLE message (Agent tool), one per area, with
model: haiku - each only reads its area and distils a card, which a lightweight model does well
at a fraction of the cost (a full-model fan-out measured ~170k tokens; haiku cuts that hard). Each
is told to search only its area, read the top 2-3 matching pages, and return a compact ready-to-use
card for its area ONLY (nothing else), citing the page paths it used:
| Agent | Searches | Returns |
|---|---|---|
| tools | wiki/tools/ |
the automated tool(s) to run + the exact command line |
| payloads | wiki/payloads/ |
ready-to-send payloads for the vuln-class |
| techniques | wiki/techniques/ |
the attack steps / chain |
| cheatsheets | wiki/cheatsheets/ |
quick copy-paste commands |
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.
- 10d ago First seen · 72 lines · 187 tokens per session scan A 28bbff0cf9c0
wiki-arsenal is a skill published in the GitHub repository Encod3d-Sec/TORCH (319 stars, last pushed 9d ago), licensed MIT. It adds 187 tokens to every session and 1,069 once invoked, about $0.0009 per session on Opus 5. 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…