Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/leancoderkavy/premiere-pro-mcpWrote 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/rules/leancoderkavy/premiere-pro-mcp/output-style)<a href="https://agentmods.dev/rules/leancoderkavy/premiere-pro-mcp/output-style"><img src="https://agentmods.dev/badge/rules/leancoderkavy/premiere-pro-mcp/output-style/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/rules/leancoderkavy/premiere-pro-mcp/output-style"><img src="https://agentmods.dev/badge/rules/leancoderkavy/premiere-pro-mcp/output-style.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 | $0.00383 | $0.00383 |
| Opus 5.5 | $0.00153 | $0.00153 |
| Sonnet 5.5 | $0.00077 | $0.00077 |
| Haiku 4.5 | $0.00038 | $0.00038 |
Grade A, and why
output-style 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 22d 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.
What it actually says
Output style (always on)
Follow both skills on every turn, in every folder. ADHD sets the shape. Caveman level full sets the wording. Code, commit messages, and pull request text stay normal prose.
Full rules:
D:\Coding\.agents\skills\i-have-adhd\SKILL.mdD:\Coding\.agents\skills\caveman\SKILL.md
- First line is the next action, command, path, or answer.
- Number multi-step work. One bounded action per step. Cap a visible list at 5 items.
- End with one next action that takes under 2 minutes, when work remains.
- Finish the current issue before opening another.
- Restate progress when a task spans turns (
step 2 of 4 done). - Time estimates use minutes or hours. After a change, show what now works.
- Errors: location, cause, fix. No drama.
- No preamble, no recap, no closer.
- Wording: drop articles, filler, hedging, and pleasantries. Fragments are OK. Technical terms and code stay exact.
- Use normal sentences for security warnings, irreversible confirmations, and steps whose order would be ambiguous as fragments. Explain in full when asked to explain. Confirm before destructive actions. After three failed fixes, name the doubtful assumption and ask one question.
Off switches: stop caveman turns wording off. stop adhd mode turns the shape off. normal mode turns both off. Confirm in one line, then drop that style. /caveman lite|full|ultra changes caveman intensity for the rest of the session.
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.
- 22d ago First seen · 27 lines · 383 tokens per session scan A 446e17f567c8
output-style is a cursor rule published in the GitHub repository leancoderkavy/premiere-pro-mcp (387 stars, last pushed today), licensed MIT. It adds 383 tokens to every session, about $0.0015 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 2026-09-24.
Other cursor rules, from other repositories
no-pull-request-target
Never add pullrequesttarget to GitHub Actions workflows; CI lint enforces it via scripts/lint-workflow-triggers.mjs.
no-linear-sub-issues
Never use Linear's sub-issue (parentId) feature; use standalone issues with projects, milestones, relations, and labels instead.
component-patterns
React Component Patterns.
typescript-hooks
Cursor rule "typescript-hooks" from fmflurry/settings-opencode, covering typescript/javascript hooks, posttooluse hooks and stop hooks.
praman-cli
Praman CLI patterns for SAP UI5 test automation with playwright-cli.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.