Is anyone else observing the error "you've hit a rate limit that restricts the number of Copilot model requests"?

[sthi-1005]

Hello, firstly thank you for the amazing tool. It's been a great help as my project grows and the code becomes more complex and chews up context super quickly.

I'm not sure if this is merely by coincidence, so I'd like to know if anyone else is observing this error since adding the context-mode prompt to the copilot-instructions.

Copilot stopped work due to an error
Sorry, you've hit a rate limit that restricts the number of Copilot model requests you can make within a specific time period. Please try again in . Please review our Terms of Service (https://docs.github.com/site-policy/github-terms/github-terms-of-service). If the problem persists, please contact GitHub Support, including the request ID #############. To retry, leave a comment on this pull request asking Copilot to try again.

This is being done through Github's web agent session.

[mksglu]

Hi @sthi-1005, thanks for reporting this!

This is not caused by context-mode — it's GitHub Copilot's own rate limiting mechanism.

Why it happens

GitHub Copilot rate limits are request-count-based, not token-based (docs). Adding context-mode instructions to copilot-instructions.md doesn't directly trigger rate limits. However, context-mode's MCP tools may lead to more tool calls per session (batch executions, searches, indexing), which naturally consumes more premium requests — reaching the limit faster.

What you can do

• Wait and retry — rate limits are temporary.
• Switch away from preview models — GitHub notes that preview models have stricter rate limits due to limited capacity.
• Check your plan's allowance — Free plans get ~50 premium requests/month, Pro gets more, Pro+ up to 1,500. If you're using the coding agent heavily, you may need a higher tier.
• Contact GitHub Support if you believe the limit is impacting legitimate use.

This is an expected trade-off: context-mode helps you use your context window more efficiently, but that efficiency comes from more granular tool calls, which count against Copilot's request quota.

Hope that helps!