Repository navigation
docs: add step-by-step setup for GitHub Copilot in VS Code - #242
Draft
Rubén (rublaman) wants to merge 2 commits into
Draft
Rubén (rublaman) wants to merge 2 commits into
Rubén (rublaman) wants to merge 2 commits into
Conversation
|
Rubén (@rublaman) please read the following Contributor License Agreement(CLA). If you agree with the CLA, please reply with the following information.
Contributor License AgreementContribution License AgreementThis Contribution License Agreement (“Agreement”) is agreed to by the party signing below (“You”),
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The Copilot installation instructions currently cover the standalone CLI. This PR adds a
Windows guide for GitHub Copilot Chat in VS Code with Copilot selected as the session target.
The existing Copilot CLI documentation stays unchanged, and both READMEs link to the new guide.
The walkthrough covers Copilot setup, downloading TokenEconomizer, installing and activating
project hooks, checking a session, and updating or removing the configuration. It includes
troubleshooting and explains which features have been checked in the editor. This setup
requires neither GitHub CLI nor standalone Copilot CLI; model calls that depend on the CLI
and unverified tool pruning remain disabled.
Testing in VS Code also exposed a parser bug: Copilot metadata containing
"tool_use"causedthe log to be read as a Claude transcript, returning no tool calls. The fix checks for actual
Claude tool blocks before falling back to Copilot events. A regression test covers successful
and failed calls, tool counts, and working directories.
Validation:
passed afterward. Following the rebase onto upstream/main, the full package suite finished
with 1168 passed and 35 skipped in 263.49 seconds on Windows. Skips cover optional integrations
and platform requirements, including symlinks and POSIX shell tests. The earlier validation
environment used Windows 11, Python 3.13.17, pytest 9.2.0, and PowerShell 7.6.10.
The full CI matrix was not run locally.
free account access. The logs reported runtime
1.0.108-0and modelgpt-6-luna. Neitherghnor
copilotwas available on the hooks' PATH. An isolated observer captured all six lifecycleevent types, successful and failed reads, and an edit. The fixed parser recovered the first
session's four calls and the final session's single read.
marker appearing in the user prompt. Closing the final test window delivered SessionEnd and
triggered automatic folding. The first end-hook observer did not finish, so that session
was folded manually. Usage totals became available after session.shutdown.
apostrophe, and a non-ASCII character. They generated six hooks, and a synthetic SessionStart
completed without a TokenEconomizer error log. The marketplace bundle built successfully;
local guide links and
git diff --checkalso passed.python -m pytest tests experiments/testsrun left 36 benchmark failures onWindows. Of 46 cases retried outside the sandbox because of Git Bash restrictions, 9 passed
and 1 skipped for Windows symlink privileges. All remaining failures were in the benchmark
suite, involving
os.getuid,signal.SIGHUP, path assertions, or shell lookup assumptions.Putting Git Bash first on PATH did not resolve every subprocess lookup. Benchmark code is unchanged.
Natural fact learning, resumed sessions, packaged plugin discovery and installation, tool
pruning, and editor-only model calls still need validation. The context-marker check verifies
injection only, and these runs did not measure token savings. The supported-version table
remains unchanged.