shotcut-mcp: Instructions file for Codex

AGENTS.md

shotcut-mcp AGENTS.md is an instructions file for Codex, OpenCode from matrodrigs/shotcut-mcp. It costs 2,191 tokens per session, scanned A, original, MIT.

Repository instructions for Shotcut MCP, a dependency-free Python server that edits Shotcut video-project files, checks them, and renders media.

In plain words
What is it for?
Use them when changing the server, its Shotcut MLT XML handling, validation, rendering, transactional writes, or dependencies.
Why use it?
They give contributors rules for protecting user projects, keeping modules maintainable, and preserving required compatibility and safety checks.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md. Also seen: mentions .claude-plugin; mentions AGENTS.md; mentions Codex.

This is matrodrigs/shotcut-mcp's own configuration. It tells Codex and OpenCode how to work on shotcut-mcp itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything shotcut-mcp configures →

Reuse

Borrowing it

Nothing to install: this file belongs to matrodrigs/shotcut-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/matrodrigs/shotcut-mcp/main/AGENTS.md
Clone the repo
git clone --depth 1 https://github.com/matrodrigs/shotcut-mcp

Made for: Codex, OpenCode.

Wrote 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.

agentmods badge for shotcut-mcp AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/matrodrigs/shotcut-mcp/agents-md/github.svg)](https://agentmods.dev/instructions/matrodrigs/shotcut-mcp/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/matrodrigs/shotcut-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/matrodrigs/shotcut-mcp/agents-md/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.

agentmods 80×15 button for shotcut-mcp AGENTS.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/matrodrigs/shotcut-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/matrodrigs/shotcut-mcp/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 2,191 This file is loaded in full into every session.
When invoked 2,191 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.02191 $0.02191
Opus 5.5 $0.00876 $0.00876
Sonnet 5.5 $0.00438 $0.00438
Haiku 4.5 $0.00219 $0.00219

Measured 21d ago against content hash 09dbcf821cf3, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

shotcut-mcp AGENTS.md 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 21d 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.

AGENTS.md · 110 lines

How it starts

The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

This file defines the engineering practices for automated and human contributors to Shotcut MCP. Apply these rules to the entire repository unless a more specific AGENTS.md exists below the file being changed.

Project intent

Shotcut MCP is a dependency-free Python MCP server that edits Shotcut MLT XML, validates it with the installed MLT runtime, renders previews and final media, and preserves user projects through transactional writes. Correctness and preservation of user data take priority over convenience or clever abstractions.

New projects target Shotcut 26.8.1 with MLT 7.41.0 serialization (7.41.x family). shotcut_mcp/__init__.py separately defines exact tested runtime pairs; the Windows CI matrix is generated from that list. To recognize a new pair, pin its archive checksum in scripts/compatibility_matrix.py and pass real integration in the same change. Unknown versions receive a warning rather than a version-based block. Keep concrete runtime failures distinct, and check RNNoise independently as both a link and a filter.

Architecture and dependency direction

Keep modules deep: expose a small interface that hides substantial behavior. Add a seam only when behavior genuinely varies or when it isolates an external dependency. Avoid pass-through wrappers, speculative ports, and circular imports.

Before changing module boundaries or imports, read the dependency graph and ownership table in docs/architecture.md. Keep that document as the architecture reference.

Preserve the public imports from project.py and platform.py. Internal modules may be reorganized without changing MCP clients, tool names, schemas, result shapes, or the documented public imports exposed by those two modules.

Non-negotiable safety invariants

  • Treat every .mlt/.xml project and rendered output as user data.
  • Never replace a project until the complete candidate passes MLT validation.
  • Recheck the project revision after validation and before replacement.
  • Use sibling temporary files, fsync where supported, and atomic replacement.
  • Preserve an existing target if validation, preview, render, or promotion fails.
  • Keep project backups isolated by canonical project identity. Restore only filenames produced by the backup store; directory membership alone is not ownership proof.
  • Preserve unknown MLT XML elements and properties unless an operation explicitly owns them. Reject duplicate IDs and ambiguous timeline roots instead of guessing.
  • Clone a shared producer before applying a clip-local mutation.
  • Treat Shotcut marker end values as exclusive. Translate a range marker to MLT's inclusive consumer out by subtracting one, and reject empty ranges.
  • Reject media replacement next to a transition instead of attempting to rewire its nested tractor. Preserve the entry range and isolate a shared producer before mutation.
  • Apply the configured allowed-root and network-resource policy to all user-controlled data paths. Executable discovery is not a data-path operation.
  • Do not invoke a shell for Shotcut, MLT, FFmpeg, or FFprobe commands. Pass argument lists directly and terminate the process group on timeout or MCP cancellation.
  • Restrict render consumer properties to single-file-safe options unless the administrator explicitly enables unsafe properties.
  • Keep MCP stdout strictly newline-delimited UTF-8 JSON-RPC. Diagnostics go to stderr.
  • Emit request progress only for a caller-supplied token, keep values strictly increasing, and omit the progress message for MCP 2024-11-05. A render token ends when start_render returns; durable progress remains owned by render_status.
  • Bound message size, in-flight work, logs, diffs, operation batches, and caches.

Read the full file on GitHub · 110 lines

Changes

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.

  1. 21d ago Changed · -75 lines · -95 tokens per session 09dbcf821cf3
  2. 1mo ago Changed · -36 lines · -334 tokens per session 36c4357e70be
  3. 2mo ago Changed · +16 lines · +200 tokens per session a5c56f85dd0e
  4. 2mo ago First seen · 205 lines · 2,420 tokens per session scan A fd96fcb26676

Subscribe to this mod's changes

shotcut-mcp AGENTS.md is an instructions file published in the GitHub repository matrodrigs/shotcut-mcp (25 stars, last pushed 5d ago), licensed MIT. It adds 2,191 tokens to every session, about $0.0088 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 2026-08-31.

Related

shotcut-mcp AGENTS.md compared with instructions from other repositories

shotcut-mcp AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
shotcut-mcp AGENTS.mdmatrodrigs/shotcut-mcpInstructions file255d agoMITA2,191Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA9,719GitHub Copilot
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA5,516GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143ktodayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149ktodayMITA4,680Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA3,093GitHub Copilot
vscode model-prompts.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA2,022GitHub Copilot