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/TJC-LP/sanzarunpx agentmods add skills/tjc-lp/sanzaru/sanzaru-cliWrote 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/tjc-lp/sanzaru/sanzaru-cli)<a href="https://agentmods.dev/skills/tjc-lp/sanzaru/sanzaru-cli"><img src="https://agentmods.dev/badge/skills/tjc-lp/sanzaru/sanzaru-cli/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/tjc-lp/sanzaru/sanzaru-cli"><img src="https://agentmods.dev/badge/skills/tjc-lp/sanzaru/sanzaru-cli.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 | $0.00088 | $0.02807 |
| Opus 5 | $0.00044 | $0.01404 |
| Sonnet 5 | $0.00018 | $0.00561 |
| Haiku 4.5 | $0.00009 | $0.00281 |
Grade A, and why
sanzaru-cli 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 — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sanzaru CLI for agents
sanzaru <group> <verb> wraps OpenAI's Sora video, gpt-image-2, TTS/Whisper, and podcast APIs
for shell use. Requires OPENAI_API_KEY in the environment. Start with:
sanzaru capabilities # no API key needed: version, enabled features, command map
Output contract (parse this, not the docs)
- stdout: exactly one JSON envelope per input —
{"v":1, "ok":true, "command":"...", "result":{...}}. Fan-out commands stream one envelope per line (JSONL) in completion order. - stderr: progress lines and hints (
sanzaru: video_x in_progress 42% t=95s). Never parse it. - Errors are envelopes too (
"ok":false,error.type, often aresumecommand) —jqnever hangs. - Exit codes:
0ok ·1runtime/API ·2usage ·3config (missing key/extra) ·4timeout — job still running, resumable ·5job failed server-side ·6partial batch ·130interrupted.
The one-shot pattern (preferred)
-o implies --download implies --wait: one command submits, polls, downloads, and prints the
final path. Run it in the background if your shell caps foreground time.
sanzaru video create "the pilot looks up and smiles" --seconds 8 --size 1280x720 \
-o ./out/pilot.mp4 --timeout 25m | jq -r .result.file.path
sanzaru image generate "an app icon, flat design" --quality high -o ./art/icon.png # sync, ~10-60s
The resume loop (harness-safe)
Submission returns in ~1s; waits are idempotent. On exit 4 the job keeps running server-side —
re-run the resume command from the envelope (or the same wait) until exit ≠ 4:
ID=$(sanzaru video create "..." --seconds 8 | jq -r .result.id)
# ...do other work, then repeatedly:
sanzaru video wait "$ID" --download -o ./out/clip.mp4 --timeout 100s
# exit 0 → done · exit 4 → re-run · exit 5 → inspect .error
sanzaru wait id1 id2 ... polls mixed video_*/resp_* ids concurrently, JSONL as each finishes.
Choosing the right image command
image generate— synchronous, RECOMMENDED for one-off images; returns file + token usage. Batch:image generate "p1" "p2" --count 2 -o ./art/(JSONL; exit 6 = partial, retry the failed.input.prompts).image create— async job; use for refinement chains:image create "add neon rain" --previous-id "$R1" -o v2.png.- gpt-image-2 is the default;
--background transparentrequires--image-model gpt-image-1.5.
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 · 174 lines · 88 tokens per session scan A 78506f24ff31
sanzaru-cli is a skill published in the GitHub repository TJC-LP/sanzaru (6 stars, last pushed yesterday), licensed MIT. It adds 88 tokens to every session and 2,807 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
webgl-holographic-foil
A self-contained WebGL2 hero: thin-film interference over a crushed-foil surface whose palette shifts with the viewing angle; move the cursor to tilt the film.
general-video
Author or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated…
html-ppt-hermes-cyber-terminal
OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.
html-ppt-taste-brutalist
16:9 HTML deck in tactical-telemetry / CRT-terminal taste. Deactivated-CRT charcoal slides, white-phosphor monospace, hazard-red accent, scanline overlay, ASCII syntax, density over decoration. Distilled from Leonxlnx/taste-skill brutalist-skill (Tactical Telemetry mode).
diagnostic-stem-delivery
Audio production with diagnostic analysis, timecode parsing from documents, and verified export workflow.
chengfeng-check-updates
An environment manager for a video-editing system. It checks whether its skills and runtime—the software needed to run them—are installed and compatible.