mcpls copilot-instructions.md

Coding guidelines for mcpls, a bridge that lets AI clients use language servers through two communication standards: MCP and LSP. Language servers provide programming features such as code understanding and diagnostics.

In plain words
What is it for?
Use it when changing mcpls architecture, asynchronous code, protocol handling, configuration, or Rust types.
Why use it?
It helps keep the bridge's Rust code type-safe, easier to maintain, and less likely to accept invalid protocol states or hide errors.

Instructions file for GitHub Copilot

Install

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.

agentmods
npx agentmods add instructions/bug-ops/mcpls/copilot-instructions
Clone the repo
git clone --depth 1 https://github.com/bug-ops/mcpls

Made for: GitHub Copilot.

Per session 1,148 This file is loaded in full into every session.
When invoked 1,148 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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 $0.01148 $0.01148
Opus 5 $0.00574 $0.00574
Sonnet 5 $0.00230 $0.00230
Haiku 4.5 $0.00115 $0.00115

Measured 2d ago against content hash 990c4ddb2006, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

mcpls copilot-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 2d 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.

.github/copilot-instructions.md · 99 lines

How it starts

The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Copilot Instructions for mcpls

mcpls is a bridge between MCP (Model Context Protocol) and LSP (Language Server Protocol). AI clients speak MCP to mcpls; mcpls spawns and communicates with language servers over LSP.

AI Client ←→ [MCP/stdio] ←→ mcpls ←→ [LSP/stdio] ←→ language server

Key crates: mcpls-core/src/bridge/, lsp/, mcp/, config/.

Type safety

Prefer newtypes over primitive aliases for domain values. A raw u32 for a line number and a raw u32 for a column are indistinguishable to the compiler; a Line(u32) and Column(u32) make transpositions a compile error.

Encode protocol state in the type system where possible. A document that has been opened and one that has not should ideally be different types, not the same type with a boolean field. Typestate prevents calling operations that are only valid on open documents.

Avoid stringly-typed dispatch. Matching on &str method names to branch protocol behaviour should be replaced with a typed enum as soon as the set of methods is known and bounded.

Option<T> and Result<T, E> must not be unwrapped with .unwrap() or .expect() in production code paths. Use ?, if let, let-else, or explicit error mapping.

Idiomatic Rust

Prefer iterator adapters over manual loops. map, filter, flat_map, collect, and fold express intent more clearly and are easier to compose than for with a mutable accumulator.

Avoid unnecessary heap allocation. Prefer &str over String in function signatures where ownership is not required. Prefer Cow<str> when a function sometimes borrows and sometimes owns.

Use let-else (stabilised in Rust 1.65) to reduce nesting when early return on None or Err is needed:

// prefer
let Some(value) = option else { return Err(...) };
// over
let value = match option { Some(v) => v, None => return Err(...) };

Derive Clone, Debug, PartialEq on data types unless there is a specific reason not to. Their absence makes testing harder and the omission is rarely intentional.

Read the full file on GitHub · 99 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. 2d ago First seen · 99 lines · 1,148 tokens per session scan A 990c4ddb2006

Subscribe to this mod's changes

mcpls copilot-instructions.md is an instructions file published in the GitHub repository bug-ops/mcpls (66 stars, last pushed 4d ago), licensed Apache-2.0. It adds 1,148 tokens to every session, about $0.0057 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 2026-08-30.