Skip to content

Instantly share code, notes, and snippets.

View T-Gro's full-sized avatar

Tomas Grosup T-Gro

View GitHub Profile
@T-Gro
T-Gro / 00-START-HERE.md
Last active October 9, 2026 13:47
Copilot skill: F# development trajectory, Microsoft team FTE, issues, PRs, T-Gro, Copilot and GH-AW attribution, and PR commit history

One prompt. A complete deck.

Load https://gist.github.com/T-Gro/1fa4da233b9345ae07c95ab59bad2cda and render evolving line charts into a slide deck.

Instructions to the receiving agent

Read the complete SKILL.md, METRICS.md, and preset.json from this gist. Use raw file URLs or the gist API if the page preview is truncated. No skill installation is required.

@T-Gro
T-Gro / spread_service_tests.diff
Last active June 12, 2026 08:47
Spread IDE service tests - RED on dotnet/fsharp#18927 (apply with git apply)
diff --git a/tests/FSharp.Compiler.Service.Tests/CompletionTests.fs b/tests/FSharp.Compiler.Service.Tests/CompletionTests.fs
index cdcbf7a0b8..048759d779 100644
--- a/tests/FSharp.Compiler.Service.Tests/CompletionTests.fs
+++ b/tests/FSharp.Compiler.Service.Tests/CompletionTests.fs
@@ -860,4 +860,31 @@ let normalize' x = x + 1
normaliz{caret}
"""
- assertHasItemWithNames [ "normalize'" ] info
\ No newline at end of file
@T-Gro
T-Gro / SpreadInlineLib.fs
Last active June 11, 2026 17:21
Conformance tests for record spreads (PR dotnet/fsharp#18927)
module SpreadInlineLib
// Library compiled to its own assembly. The inline body below is serialized
// into the assembly's pickled TypedTree and re-elaborated at the caller's
// site in another assembly (Spreading_v1.fsx), exercising the spread
// elaboration across the TypedTreePickle boundary.
type Lbl = { A : int; B : int }
let inline bump (x: Lbl) = { ...x; A = 99 }
@T-Gro
T-Gro / employee-profile-extraction.md
Last active April 21, 2026 13:45
Employee Profile Extraction Pipeline — for Copilot CLI Opus sessions
name employee-profile-extraction
description Extracts a generalized employee profile from the user's full Copilot CLI session history. Invoke to understand who someone is, what they do, and how they work — not what they did last week.

Employee Profile Extraction Pipeline

Build a generalized employee profile from Copilot CLI local session data. Produces a structural description of role, work inboxes, competencies, and work patterns — like an employee description, not a sprint report.

Pipeline

@T-Gro
T-Gro / profile-summary.md
Created April 21, 2026 13:37
Employee profile — extracted from 3 weeks of Copilot CLI session data

Senior F# Compiler Engineer & Developer Toolsmith

Microsoft (.NET Division) — F# Compiler & Tooling Team

Compiler engineer who builds the tools that build the compiler — owns deep type-system and codegen work in a major open-source compiler while pioneering AI-augmented development workflows for the team.

Work Inboxes

Inbox Source Shape Processing
@T-Gro
T-Gro / profile-summary.md
Created April 21, 2026 13:37
Employee profile — extracted from 3 weeks of Copilot CLI session data

Senior F# Compiler Engineer & Developer Toolsmith

Microsoft (.NET Division) — F# Compiler & Tooling Team

Compiler engineer who builds the tools that build the compiler — owns deep type-system and codegen work in a major open-source compiler while pioneering AI-augmented development workflows for the team.

Work Inboxes

Inbox Source Shape Processing
@T-Gro
T-Gro / expert-reviewer-extraction-pipeline.md
Created March 17, 2026 10:32
Expert Reviewer Extraction Pipeline - generate instructions, skills and review agent from GitHub review history

Expert Reviewer Extraction Pipeline

Generate folder-scoped instructions, topic-scoped skills, and a multi-dimensional review agent from a GitHub user's public review history. Produces anonymized, deduplicated, Copilot-compatible .github/ artifacts.

Scale Warning

This pipeline processes thousands of GitHub items (typically 3,000–10,000+ issues, PRs, discussions, and review comments spanning a decade). It will not fit in a single context window.

Use sub-agents for everything. The orchestrator manages SQLite state and dispatches work. Sub-agents do the heavy lifting:

  • Collection: one sub-agent per repo × date range chunk. Parallelize aggressively (6+ concurrent agents).
@T-Gro
T-Gro / ffs-mcp.ts
Last active October 22, 2025 11:34
FFS MCP server
#!/usr/bin/env -S deno run --allow-read --allow-write --allow-env
import { Server } from "npm:@modelcontextprotocol/sdk@0.5.0/server/index.js";
import { StdioServerTransport } from "npm:@modelcontextprotocol/sdk@0.5.0/server/stdio.js";
import {
CallToolRequestSchema,
ListToolsRequestSchema,
} from "npm:@modelcontextprotocol/sdk@0.5.0/types.js";
const server = new Server({
name: "ffs-instructions",