github / github/copilot-cli

Do not derive model context from prompt and output token limits

Open Beginner friendly
#4,638 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

triage
Dominant language
Shell
Stars
11.2k
Forks
1.9k
Avg merge
14h 16m
Merged PRs (30d)
6

Description

Describe the bug

The model details display can overstate total context if it derives context by adding max_prompt_tokens and max_output_tokens. These are independent maxima and their sum may exceed max_context_window_tokens.

Affected version

1.0.81

Steps to reproduce the behavior
  1. Return synthetic model metadata with max_prompt_tokens: 100000, max_output_tokens: 20000, and max_context_window_tokens: 100000.
  2. Open the CLI model details display.
  3. Check whether it reports 120,000 tokens instead of the declared 100,000-token context window.
Expected behavior

Display max_context_window_tokens as the total context window. Treat prompt and output limits as independent maxima rather than values whose sum defines context.

Additional context

The Copilot SDK/runtime preserves the independent context value, so this appears limited to CLI presentation.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start at the CLI model details display and trace how synthetic model metadata is formatted. Reproduce with max_prompt_tokens 100000, max_output_tokens 20000, and max_context_window_tokens 100000; done means the display reports the declared 100000-token context window while keeping prompt and output limits independent.

Written by the indexing model from the issue text.

Assessment

Tech stack
shell
Domain
cli
Issue type
Bug
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Active
Clarity
Clearly specified
Newbie friendliness
75/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.