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/screenshotWrote 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/screenshot)<a href="https://agentmods.dev/skills/encod3d-sec/torch/screenshot"><img src="https://agentmods.dev/badge/skills/encod3d-sec/torch/screenshot/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/screenshot"><img src="https://agentmods.dev/badge/skills/encod3d-sec/torch/screenshot.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high YARA Match · line 119 YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).Fix: Remove offensive tool references and exploit code. Legitimate agent skills should not contain penetration testing tools, exploit frameworks, or reconnaissance utilities.
- high YARA Match · line 128 YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).Fix: Remove offensive tool references and exploit code. Legitimate agent skills should not contain penetration testing tools, exploit frameworks, or reconnaissance utilities.
- medium Data Exfiltration · line 54 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
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 | $0.00099 | $0.03875 |
| Opus 5.5 | $0.00040 | $0.01550 |
| Sonnet 5.5 | $0.00020 | $0.00775 |
| Haiku 4.5 | $0.00010 | $0.00387 |
Grade A, and why
screenshot scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
is fine (it's a GET you already make). Never screenshot a host you would not curl. How it starts
The opening of the file, as written. The whole thing — 208 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Screenshot: visual PoC capture
Turn the text walkthrough into evidence. Capture runs on the Kali tooling host (it has the VPN
route to in-scope targets); the PNG is then pulled into the vault under targets/<eng>/poc/. Driver:
scripts/shot.py (chromium headless). For redaction-before-report use Skill(evidence).
One evidence dir, captured by hand. poc/ = the curated story you choose to show (login, the
exploited state, the flag). ALL PoC evidence is now captured MANUALLY and LIVE via capture.sh straight
into poc/ the moment a step lands - there is no auto-capture net.
Scope check
Only screenshot in-scope hosts (read targets/<eng>/scope.md). On passive_only RoE a screenshot
is fine (it's a GET you already make). Never screenshot a host you would not curl.
What to capture: EVERY breakthrough, as it happens (with retries)
Screenshot the moment each step LANDS, not at the end - the box may expire. A breakthrough = anything
you'd put in the report: the entry page, a leaked secret / useful source (deobfuscated JS, a
config, an app.log), the cipher / decrypt response, the vuln firing, the authed/exploited
state, the flag. Tell the arc (Entry -> Trigger -> Impact) AND capture the intermediate proofs
(the leak, the crypto response) - those are exactly what a reviewer disbelieves without a picture.
- Always show the URL. Every web/source capture must carry its address bar so the image is
self-identifying - which page/path it is (
https://T/folder/folder/log.md), not an anonymous blob. Live<url>mode adds it automatically; for--html/--termof a fetched resource pass--url-bar "<full URL>". - Retry (box is up). Renders fail transiently (nav timeout, cold chromium). Retry 2-3x and confirm
the PNG is non-empty (
[ -s out.png ]) before moving on; while the box is up, go back and re-capture anything you missed rather than shipping evidence with gaps. NameNN-slug.png(NN = step order) so the evidence reads top-to-bottom.
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 · 208 lines · 99 tokens per session scan A 2d3c87c61457
screenshot is a skill published in the GitHub repository Encod3d-Sec/TORCH (336 stars, last pushed 1mo ago), licensed MIT. It adds 99 tokens to every session and 3,875 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). 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…