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/perso-ai/perso-dubbing-pluginnpx agentmods add skills/perso-ai/perso-dubbing-plugin/srtWrote 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/perso-ai/perso-dubbing-plugin/srt)<a href="https://agentmods.dev/skills/perso-ai/perso-dubbing-plugin/srt"><img src="https://agentmods.dev/badge/skills/perso-ai/perso-dubbing-plugin/srt/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/perso-ai/perso-dubbing-plugin/srt"><img src="https://agentmods.dev/badge/skills/perso-ai/perso-dubbing-plugin/srt.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
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 →
- medium Output Handling · line 78 Output size or generation rate is not bounded. Unbounded output enables denial-of-service through resource exhaustion, log flooding, or context-window stuffing.Fix: Set explicit limits on output length, generation count, and rate. Use max_tokens and truncation to prevent unbounded output.
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 | $0.00027 | $0.04188 |
| Opus 5.5 | $0.00011 | $0.01675 |
| Sonnet 5.5 | $0.00005 | $0.00838 |
| Haiku 4.5 | $0.00003 | $0.00419 |
Grade A, and why
srt 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 — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/srt
A skill that extracts the source-language subtitles of a video/audio via the Perso AI STT API — and then you (the agent) translate them into the requested languages.
Core rules (must follow)
- Only the worker sees the raw key. Never open the key with
Read, echo it, or pass it as a command-line argument. Workers send it in theXP-API-KEYheader only. - One command = the whole extraction.
scripts/srt.mjshandles upload · transcription · SRT download by itself. - Always run
srt.mjsin the background. Transcription takes minutes; a foreground shell timeout kills the run mid-way. - Language unspecified → target
en. - Transcription only, no translation (user asks for "the subtitles/transcript as-is", "just the SRT in the original language") → run with
--transcribe-onlyinstead of--target. The downloaded original SRT is the final deliverable — skip the Translate section entirely. - Multiple languages go in one command (
--target en,ja) — one subtitle project is created per input regardless of how many languages, and you translate its single original SRT into each of them. - Relay progress faithfully. Surface the worker's stdout
[progress]lines to chat, verbatim or summarized. The indented stderr detail logs don't need relaying. For a batch the worker streams a[progress]line per input and prints a[status]heartbeat (subtitles 3/20, with anext check ~Nmhint) — check the background output ~3 minutes after launch, then on the hinted interval, relaying new lines. Never invent percentages or ETAs. - Name your runtime for telemetry (silent). Add
--host <claude-code|codex|cursor|antigravity|gemini>to the worker command, naming the agent you are running as, if you can tell. It only tags usage telemetry — never surface it to the user and never ask them about it. Omit it if unsure. - Unsupported formats are skipped automatically and the rest keep processing — relay the skip notice.
- Offline operations never need a key — never run any key check/registration for them:
srt.mjs --check/--retime,style.mjswith a local video + SRT (everything except--project), and translating an SRT the user hands you. - This skill shares the dubbing skill's libraries: the
dubbingfolder must be installed next tosrt(every install method ships them together; only a manual copy of thesrtfolder alone breaks this).
What ships with it
8 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 145 lines · 27 tokens per session scan A e93c5fc91895
srt is a skill published in the GitHub repository perso-ai/perso-dubbing-plugin (40 stars, last pushed 12d ago), licensed MIT. It adds 27 tokens to every session and 4,188 once invoked, about $0.0001 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
chinese-documentation
A Chinese technical-documentation style guide covering spacing, punctuation, numbers, terminology, and links when Chinese and English appear together.
oma-translation
Translate or revise localized UI strings and prose while preserving meaning, terminology, placeholders, and structure.
speak-human-tw
A Traditional Chinese editing guide for making public-facing writing sound more natural and less machine-generated. It also checks wording associated with Mainland China and half-width punctuation.
page-metadata
When the user wants a full-page metadata audit or implementation across discovery, search intent, title, description, canonical, robots, hreflang, Open Graph, Twitter Cards, code or CMS changes, build validation, rendered HTML, production verification, localization, or post-release observation. Also use for "metadata…
localization-strategy
Plan, audit, implement, migrate, or validate localization for multilingual products and websites, including locale scope, market readiness, URL architecture, content coverage, hreflang coordination, rollout, and production verification. Use translation for the actual translation, glossary, and language review work.
translation
Translate, transcreate, review, or maintain multilingual product and marketing content, including source locking, glossary decisions, language style, machine-assisted workflows, legal-content safeguards, QA, and source-change synchronization. Use localization-strategy for market scope, locale architecture, rollout…