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.
npx agentmods add instructions/matrodrigs/shotcut-mcp/agents-mdgit clone --depth 1 https://github.com/matrodrigs/shotcut-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/instructions/matrodrigs/shotcut-mcp/agents-md)<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="Measured on agentmods" 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 | $0.02420 | $0.02420 |
| Opus 5 | $0.01210 | $0.01210 |
| Sonnet 5 | $0.00484 | $0.00484 |
| Haiku 4.5 | $0.00242 | $0.00242 |
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 5d 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 — 205 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.
The validated compatibility target is Shotcut 26.6.25 with MLT 7.40.0 serialization in the
MLT 7.40.x family. The runtime source of truth is shotcut_mcp/__init__.py. Do not infer
that a successful MLT repository preflight proves an optional service loaded; RNNoise
must continue to be checked 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.
The intended dependency direction is:
server -> protocol, tools
tools -> platform, project, protocol, render
project -> missing_media, platform, project_document, project_snapshot, protocol, storage
missing_media -> platform, protocol
project_document -> media, mlt_xml
project_snapshot -> project_document, mlt_xml, path_policy
render -> platform, project_snapshot, protocol, render_jobs, storage
render_worker -> platform, render_jobs, storage
platform -> media, path_policy, processes, protocol, storage
media -> processes, protocol
path_policy -> mlt_xml
processes -> path_policy, protocol
Module ownership:
server.py: JSON-RPC/MCP lifecycle, concurrency, cancellation, progress transport, and wire compatibility.tools.py: MCP tool catalog, schemas, annotations, and thin handler routing.project.py: public project-level orchestration and transaction workflow.missing_media.py: bounded missing-resource discovery, scoring, and visualization.project_document.py: MLT XML model, timeline invariants, and edit semantics.project_snapshot.py: read-only MCP projection and stable timing facts from an MLT document.platform.py: stable public orchestration interface for Shotcut and MLT operations.path_policy.py: canonical path resolution and embedded network-resource policy.processes.py: executable discovery and cancellable child-process supervision.media.py: FFprobe execution, caching, normalized media summaries, and bounded parsing of FFmpeg quality analyzers.mlt_xml.py: shared decoding of MLT properties and clock values.render.py: public durable render-job lifecycle.render_worker.py: out-of-process render ownership and final output promotion.render_jobs.py: private persistent job metadata and bounded logs.storage.py: locks, backups, revisions, and atomic output transactions.protocol.py: schema validation plus request-scoped cancellation and progress context.
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.
- 5d ago First seen · 205 lines · 2,420 tokens per session scan A fd96fcb26676
shotcut-mcp AGENTS.md is an instructions file published in the GitHub repository matrodrigs/shotcut-mcp (3 stars, last pushed yesterday), licensed MIT. It adds 2,420 tokens to every session, about $0.0121 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 instructions, from other repositories
ghidra-mcp CLAUDE.md
Claude Code instructions for bethington/ghidra-mcp, covering ghidra mcp - claude code guide, boil the ocean, community interaction (read-only by default), repository scope — what belongs here and architecture.
ghidra-mcp AGENTS.md
AGENTS.md instructions for bethington/ghidra-mcp, covering agents.md — ghidra-mcp project, project context, directory structure, current priorities and guidelines.
unraid CLAUDE.md
Claude Code instructions for dinglebear-ai/unraid, covering claude.md, what this is, identity and history (read before touching any url), "unraid-mcp" means five different things and agent plugins ship no claude code hooks.
Cortex CLAUDE.md
Claude Code instructions for cdeust/Cortex, covering cortex — persistent memory mcp server, problem statement, build & test, releasing and architecture.
wet-mcp AGENTS.md
AGENTS.md instructions for n24q02m/wet-mcp, covering agents.md - wet-mcp, cau truc, lenh thuong dung, mise shortcuts and cau hinh quan trong.
servicenow-platform-mcp AGENTS.md
AGENTS.md instructions for Xerrion/servicenow-platform-mcp, covering agents.md - servicenow-platform-mcp, 📋 project overview, dependencies, 🚀 setup and 🔧 lint / format / type-check.