Borrowing it
Nothing to install: this file belongs to sbroenne/mcp-server-powerpoint. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/sbroenne/mcp-server-powerpoint/main/.github/instructions/critical-rules.instructions.mdgit clone --depth 1 https://github.com/sbroenne/mcp-server-powerpointWrote 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/sbroenne/mcp-server-powerpoint/critical-rules)<a href="https://agentmods.dev/instructions/sbroenne/mcp-server-powerpoint/critical-rules"><img src="https://agentmods.dev/badge/instructions/sbroenne/mcp-server-powerpoint/critical-rules/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/instructions/sbroenne/mcp-server-powerpoint/critical-rules"><img src="https://agentmods.dev/badge/instructions/sbroenne/mcp-server-powerpoint/critical-rules.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 | $0.00589 | $0.00589 |
| Opus 5.5 | $0.00236 | $0.00236 |
| Sonnet 5.5 | $0.00118 | $0.00118 |
| Haiku 4.5 | $0.00059 | $0.00059 |
Grade A, and why
mcp-server-powerpoint critical-rules.instructions.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 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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Critical rules
Rule 1: Success and errors
Success == true requires ErrorMessage == null. Set success only on the
success path. Expected caller-correctable failures (bad indices, missing files,
unknown sessions) return Success=false; validate before the failing COM call.
Rule 1b: Exception propagation
Never catch exceptions from batch.Execute() in Core and turn them into error
results. Unexpected COM/runtime exceptions propagate through the batch.
PowerPointToolsBase.ExecuteToolAction is the MCP catch-all boundary: it logs
the HResult to stderr and returns a structured error. Do not add another
catch-all in individual tools. MCP stdout is reserved for JSON-RPC.
COM access and ownership
- All PowerPoint access runs inside
batch.Executeon its STA thread. - Use typed PowerPoint/Office PIA members and enums. Before using
dynamic, reflection, raw dispatch, or numeric enum values, confirm that the restored interop metadata lacks the typed API, document the missing surface, and cover the exception with a real-COM test. Convenience is not an exception. - Release every manually acquired COM reference in
finally, including typed references. Embedded PIA types do not manage COM reference lifetimes. - Slide, shape, row, and column indices are 1-based. Zero and negative values are validation failures, not alternative indexing.
Rule 30: Behavior changes need real evidence
Write a focused failing regression test, observe the expected failure, then implement and rerun it. Never mock COM for Core commands: use real PowerPoint. Keep COM tests serialized and use explicit execution timeouts. Protocol-only MCP tests may use the SDK in-memory transport. See testing strategy for filters and scope.
Session lifecycle
- Track every live session in
PresentationSessionRegistry. On host shutdown, bothPresentationSessionShutdownService.StopAsyncandMain'sfinallycall the idempotentDisposeAll()backstop. - Close removes a session immediately and disposes its batch in the background; do not block close on process exit. PowerPoint may take minutes to exit after Quit. Do not force-kill on the happy path.
- Cleanup targets only proven process ownership (PID plus start time), never process names, window-title matches, or another user's PowerPoint.
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 Changed · -75 lines · -897 tokens per session fd806bcdf650
- 2mo ago First seen · 135 lines · 1,486 tokens per session scan A b190cc8cb438
mcp-server-powerpoint critical-rules.instructions.md is an instructions file published in the GitHub repository sbroenne/mcp-server-powerpoint (37 stars, last pushed yesterday), licensed MIT. It adds 589 tokens to every session, about $0.0024 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 instructions, from other repositories
mcp-server-excel copilot-instructions.md
Copilot instructions for sbroenne/mcp-server-excel: This repository contains generated plugin publication output. Before changing anything, read the repository warning.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).