EcoCode Insights VS Code Extension

EcoCode Insights is a VS Code extension that runs EcoCode CLI scans and displays repository performance, energy/runtime metrics in a dashboard.

Screenshots

Inline optimization suggestions

Energy/performance issues are flagged as you type — inline diagnostics (squiggles) with severity by impact, plus a panel for the current file. Each suggestion carries its rule ID and line number.

Optimization suggestions and inline diagnostics

Workspace dashboard

Totals, average sustainability score and the collector in use. Workspace scans use the safe static estimate by default (it never executes your files).

Workspace summary dashboard

Measured vs Estimated, per file

Every file is clearly badged, so an estimate is never mistaken for a real measurement.

Top files with measured/estimated badges

Current file at a glance

CPU, memory, energy and sustainability score for the file you are editing, alongside the dashboard.

Current file metrics with estimated badge

Run-to-run stability

Repeated runs expose variability (coefficient of variation) so you can trust the numbers before gating on them.

Stability panel

Installation

The extension is a UI on top of the EcoCode CLI (a Python package). Install the CLI once — either:

One click: run EcoCode: Setup CLI In Workspace (or click Setup CLI when prompted). It uses pipx when available, otherwise a dedicated virtual environment — PEP 668-safe on Linux/macOS/Windows, and the extension auto-detects either install.
Manually with pipx:
```
pipx install ecocode-cli
```
(No pipx? sudo apt install pipx or python3 -m pip install --user pipx, then pipx ensurepath.) On Debian/Ubuntu/WSL, a plain pip install into the system Python is blocked by PEP 668 — use pipx or a venv.

Requires Python 3.10+. If the CLI is installed in a custom location, set ecocode.cliPath to its absolute path. Repo: https://github.com/LeonardLeroy/EcoCode

AI suggestions are optional (no API key needed)

EcoCode works fully offline with deterministic, rule-based suggestions — the inline squiggles need no API key and no network.

AI-powered suggestions are an opt-in extra, configured in ecocode.toml:

Local (Ollama): provider = "ollama" — runs a model on your machine, your code never leaves it, no key.
Remote (Anthropic): provider = "anthropic" — higher quality, but your source is sent to the API, so it requires your own API key via the ECOCODE_LLM_API_KEY environment variable (never stored in settings or the repo).

See the project documentation for details.

Features

Scan whole workspace with EcoCode CLI.
Scan currently opened file.
Auto refresh scans on interval.
Show summary, stability, top files, and current-file metrics.
Inline optimization suggestions as diagnostics (squiggles) while you edit, with severity by impact.
Quick-fixes (code actions) for patchable Python rules, plus an LLM-generated candidate — always shown as a diff before applying.
Suggestions panel in the dashboard for the current file.
Measured vs Estimated badges so energy figures are never misleading.
Configure scan filters directly from command palette.

Measured vs Estimated

Workspace scans use the static collector by default: a source-based estimate that never executes your files (safe and fast). The runtime collector performs a real measurement by executing a single file — it supports Python and interpreted languages (Node, Ruby, PHP, Bash, …) when the interpreter is on PATH. Each result is labelled measured or estimated in the dashboard.

Commands

EcoCode: Open Dashboard
EcoCode: Scan Workspace
EcoCode: Scan Current File
EcoCode: Suggest Optimizations For Current File
EcoCode: Start Auto Refresh
EcoCode: Stop Auto Refresh
EcoCode: Configure Scan Filters

Settings

ecocode.cliPath: path to ecocode executable.
ecocode.collector: static (default, no execution), runtime (real measurement), or placeholder (synthetic).
ecocode.diagnosticsEnabled: show inline optimization suggestions while editing (default true).
ecocode.timeoutSeconds: max seconds per CLI invocation (default 120).
ecocode.installSource: pip install source used by Setup CLI (PyPI name, git URL, or local path).
ecocode.maxFiles: max number of files scanned per workspace scan (default 200). Large repos are capped at this number; when the limit is reached the dashboard shows a "showing X of Y files" banner with a one-click Increase limit button. Raise it to scan more.
ecocode.runs: repeated runs for stability.
ecocode.extensions: optional extension filters.
ecocode.includeGlobs: optional include globs.
ecocode.excludeGlobs: optional exclude globs.
ecocode.autoRefreshSeconds: auto refresh interval.
ecocode.showTopFiles: max files shown in dashboard table.

Local Development

cd vscode-extension
npm install
npm run compile

EcoCode Insights

Leonard Leroy