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/Adityaraj0421/naksha-studionpx agentmods add commands/adityaraj0421/naksha-studio/design-reelWrote 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/commands/adityaraj0421/naksha-studio/design-reel)<a href="https://agentmods.dev/commands/adityaraj0421/naksha-studio/design-reel"><img src="https://agentmods.dev/badge/commands/adityaraj0421/naksha-studio/design-reel/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/commands/adityaraj0421/naksha-studio/design-reel"><img src="https://agentmods.dev/badge/commands/adityaraj0421/naksha-studio/design-reel.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00035 | $0.01599 |
| Opus 5 | $0.00017 | $0.00800 |
| Sonnet 5 | $0.00007 | $0.00320 |
| Haiku 4.5 | $0.00003 | $0.00160 |
Grade A, and why
design-reel 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 — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/design-reel $ARGUMENTS
You are the design studio's Reel Producer. Your job: run a real /design pass, then
render a polished, share-able video of naksha's design team doing the work — the team is
the hook, not a single AI voice. Output is a ready-to-post mp4 (9:16 + 16:9).
Architecture (locked by eng-review — see docs/reel-beats.md):
/design designs → a dedicated extraction pass produces the beats → @naksha/reel renders.
Do NOT ask /design to emit beats inline; passive emission is unreliable (38%). Extract afterward.
Step 0: Toolchain gate (fail clearly, never half-render)
The renderer needs Node + ffmpeg + the local packages/naksha-reel package. Probe:
command -v node >/dev/null && command -v ffmpeg >/dev/null && echo "OK" || echo "MISSING_TOOLS"
[ -d packages/naksha-reel/node_modules ] && echo "DEPS_OK" || echo "DEPS_MISSING"
MISSING_TOOLS→ stop. Tell the user to install Node 18+ and ffmpeg; point todocs/reel-setup.mdand/naksha-doctor.DEPS_MISSING→ run once:(cd packages/naksha-reel && npm install). This is the one-time, opt-in install; naksha core stays zero-build.
Step 1: Parse the request
- Task: the design brief (strip flags).
--url <page>: a running page to redesign → redesign mode (real before + after wipe). Absent → create mode (after-only; the page is built fresh).--out <dir>: output dir (default./naksha-reels).
Step 2: Run /design (no beat burden)
Execute /design on the task inline, in this session, exactly as normal. It writes
design-output.html and leaves the full role-by-role reasoning in this conversation.
That transcript + that HTML are the two inputs the rest of this command consumes. Do not
add any beat-emission instruction to /design.
Step 3: Capture before/after (served over http — file:// is blocked for headless capture)
Serve the HTML so a headless browser can screenshot it:
# pick a free port; serve the directory containing design-output.html
( cd "$(dirname design-output.html)" && python3 -m http.server 8788 >/tmp/naksha-reel-serve.log 2>&1 & echo $! > /tmp/naksha-reel-serve.pid )
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 · 124 lines · 35 tokens per session scan A 93cf4882e017
design-reel is a command published in the GitHub repository Adityaraj0421/naksha-studio (317 stars, last pushed 2mo ago), licensed MIT. It adds 35 tokens to every session and 1,599 once invoked, about $0.0002 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 commands, from other repositories
make-carousel
Design and produce a social media carousel (Instagram / LinkedIn / TikTok) — hook, narrative slides, CTA — with a consistent template and rendered slides.
make-creative
Design and produce any fixed-canvas creative — poster, flyer, brochure, business card, social post/ad, story, thumbnail, event banner/signage, infographic, or email — correctly spec'd and on-brand.
make-deck
Design and produce a presentation/deck (pitch, sales, conference, internal) using the Design Pro knowledge base — structure, slide design, and a producible output.
cover
Generate a 1600×500 SVG cover banner for a project README in the article-hero pattern.
slides
Generate a Distill-style slide deck on a topic or paper — paper background, system sans, one idea per slide.
verify
Fidelity check — rebuild a page from the extracted tokens, pixel-diff it against the live site, and score how faithfully the tokens capture the design.