Raise Repo Assist fallback AI credit cap to prevent guardrail-triggered agent aborts

[Copilot]

Repo Assist scheduled runs are failing when usage narrowly exceeds the default AI credit guardrail (1000), causing the Copilot engine to terminate before completing work. This PR increases Repo Assist’s fallback cap to add headroom while preserving the same guardrail mechanism.

• Workflow guardrail adjustment

  • Updated repo-assist.lock.yml to increase the fallback GH_AW_MAX_AI_CREDITS from 1000 to 1200.
  • Applied the change in both places where the workflow sets/propagates the fallback value so runtime config and downstream env are consistent.

• Behavioral impact

  • Keeps org/repo variable override behavior unchanged (vars.GH_AW_DEFAULT_MAX_AI_CREDITS still takes precedence).
  • Only changes the fallback used when that variable is unset.

# .github/workflows/repo-assist.lock.yml
GH_AW_MAX_AI_CREDITS="${{ vars.GH_AW_DEFAULT_MAX_AI_CREDITS || '1200' }}"
GH_AW_MAX_AI_CREDITS: ${{ vars.GH_AW_DEFAULT_MAX_AI_CREDITS || '1200' }}

[Copilot]

Pull request overview

This PR increases the fallback AI credit cap used by the Repo Assist scheduled workflow to reduce guardrail-triggered aborts when usage slightly exceeds the current default, while keeping the existing vars.GH_AW_DEFAULT_MAX_AI_CREDITS override behavior unchanged.

Changes:

• Raised the fallback GH_AW_MAX_AI_CREDITS from 1000 to 1200 in the agent step where the AWF config is generated.
• Raised the same fallback value in the downstream env: propagation to keep runtime config and job environment consistent.

Show a summary per file

File	Description
.github/workflows/repo-assist.lock.yml	Bumps the workflow’s fallback max AI credits from 1000 → 1200 in both the runtime assignment and propagated env var.

Copilot's findings

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

• Files reviewed: 1/1 changed files
• Comments generated: 0