Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Encod3d-Sec/TORCHnpx agentmods add skills/encod3d-sec/torch/walkthroughWrote 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/walkthrough)<a href="https://agentmods.dev/skills/encod3d-sec/torch/walkthrough"><img src="https://agentmods.dev/badge/skills/encod3d-sec/torch/walkthrough/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/walkthrough"><img src="https://agentmods.dev/badge/skills/encod3d-sec/torch/walkthrough.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 | $0.00103 | $0.01396 |
| Opus 5.5 | $0.00041 | $0.00558 |
| Sonnet 5.5 | $0.00021 | $0.00279 |
| Haiku 4.5 | $0.00010 | $0.00140 |
Grade A, and why
walkthrough 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 1mo 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 — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Walkthrough auto-assembly
Turn a solved engagement into the full, report-ready targets/<eng>/walkthrough.md: rendered
evidence, a populated ## Evidence gallery, and a drafted narrative, so the operator only
reviews/polishes instead of assembling from scratch.
Convention: mark close-out explicitly
At close-out, write a STATUS heading into state.md:
## STATUS: SOLVED
(OWNED / ROOTED / COMPLETE also count.) This is the close-out signal: once present, the
CLAUDE.md execution loop runs Skill(walkthrough) (then Skill(learn)).
Steps
(a) Confirm the evidence is on disk (capture any missing key state now)
Evidence is captured LIVE during the engagement, straight into poc/ (via capture.sh /
Skill(screenshot)) -- there is no staging/drain step anymore. Confirm the PNGs are on disk,
and if a key state (foothold shell, the flag, an exploited render) was never captured, capture
it now before assembling:
ls targets/<eng>/poc/*.png targets/<eng>/poc/**/*.png 2>/dev/null
# missing a key state? capture it live, e.g.:
# bash scripts/capture.sh ev <eng> <slug> <url> "<cmd-label>"
A walkthrough with an empty gallery means evidence was not captured as steps landed -- fix that by capturing the reproducible states now, not by fabricating.
(b) Scaffold + gallery
python3 scripts/build-walkthrough.py <eng>
Idempotent: scaffolds the walkthrough structure from the framework template if missing, and
populates the ## Evidence gallery from every rendered card on disk. Never clobbers existing
narrative -- safe to re-run after step (a).
(c) Draft the narrative -- never fabricate
Read state.md (for ctf, also its ## Chain/## Status sections, the live working copy of
the attack path and the SOLVED/flags marker) and loot.md for the active engagement; for
pentest/bugbounty also read Killchain.md and log.md (a ctf engagement has neither; its live
chain and narrative live in state.md instead). Write the step-by-step reproduction into the
non-Evidence sections (Access -> Recon -> Foothold -> Privilege escalation -> root/flag), using
the EXACT commands, creds, and per-step results already captured in those files. If a fact needed
for a section is not present in the state files, do NOT invent it; leave a clearly marked
_TODO: <what is missing>_ for the operator instead.
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.
- 1mo ago First seen · 95 lines · 103 tokens per session scan A f03c85ba3fde
walkthrough is a skill published in the GitHub repository Encod3d-Sec/TORCH (336 stars, last pushed 1mo ago), licensed MIT. It adds 103 tokens to every session and 1,396 once invoked, about $0.0004 per session on Opus 5.5. 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
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…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
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…