Use case · Free plugin
I asked my agent to sort 40 new issues. Now the model reads the groups and the counts.
The free context-mode plugin lets your agent read a full issue list, a pull request list or hundreds of commits on your machine. A short script groups and counts them. On the Vite repository, 40 GitHub issues as JSON came to 178.7 KB, and the model read 0.8 KB.
Full list: every body, label and comment reaches the model.
Every later call sends it again.
Free plugin: a script groups and counts the issues on your machine.
Sent to the model
- The groups by area, the likely duplicates and three picks
Kept on your machine
- The full JSON, searchable with ctx_search
Each issue brings its body, labels and comments. The list gets big.
You ask your agent to sort them by area, find the duplicates and pick the three to fix first.
With bodies, labels, comments and reactions, the list is larger than most source files.
Check runs and git log --stat over hundreds of commits grow the same way.
A script sorts the list. The model reads the result.
- 01The command runs in the sandbox
The agent runs
ghorgitthroughctx_execute, on your machine. - 02A short script groups and counts
It sorts the items by area, label or file. The model reads its print.
- 03The full list stays searchable
The JSON stays in the local index. When you ask about one issue later, the agent searches for it and reads that issue.
Three lists, measured. The model reads under 1 KB of each.
| List | Full output | The model read | Time |
|---|---|---|---|
GitHub issues (gh issue list --json, vitejs/vite, 40 items) | 178.7 KB | 0.8 KB | 65 ms |
GitHub pull requests with checks (gh pr list --json, vitejs/vite, 25 items) | 190.9 KB | 0.8 KB | 79 ms |
Commit history with file stats (git log --stat, 300 commits, vitejs/vite) | 160.5 KB | 0.3 KB | 66 ms |
Plugin benchmark of 2026-11-01 on an Apple M1 Pro, default settings, median of 5 runs. Raw data: the result tables. Rerun it with the steps on Benchmarks.
Give your agent one of these.
List the open issues labelled bug. Group them by area, mark the likely duplicates, and tell me which three to fix first and why.
Read the last 300 commits with file stats. Which files change together most often, and which areas changed the most this month?
Check the open pull requests. Which ones have red checks, what is the first error in each, and which ones are ready to merge?
What this uses.
Open source. It runs in your agent, on your machine.
Where gh and git run and what comes back.
Your GitHub token stays in your own gh setup.
Sort the whole list. Read the short answer.
Each agent installs the free plugin with its own install command.