claude anthropic developer-tools billing-issues

HERMES.md in commit messages causes requests to route to extra usage billing

HERMES.md in git commit messages causes requests to route to extra usage billing instead of plan quota · Issue #53262 · anthropics/claude-code

github.com

April 29, 2026

2 min read

🔥🔥🔥🔥🔥

77/100

Summary

Including the case-sensitive string HERMES.md in git commit messages causes Claude Code to route API requests to "extra usage" billing instead of the plan's quota. This misrouting can lead to unexpected charges, as demonstrated by a $200 increase in extra usage credits while remaining within the Max 20x plan capacity.

Key Takeaways

The presence of "HERMES.md" in git commit messages causes Claude Code to route API requests to extra usage billing instead of the included plan quota.
Users can incur significant extra charges, as demonstrated by a case where $200 in extra usage credits were consumed due to this issue.
The error message received when exceeding extra usage does not indicate that the commit message content is the cause, complicating diagnosis for affected users.
API request billing should not depend on the content of git commit messages, as it leads to unexpected billing for users on the Max plan.

Read original article

View story →

Coverage & discussions

Hacker News · github.com

HERMES.md in commit messages causes requests to route to extra usage billing ↗

Community discussions

Hacker News discussion ↗

Historical count: 5 · 5 saved in sample · Historical sample

Analysis ready

Community Sentiment

Mixed

Positives

Anthropic's prompt response to the billing issue, offering full refunds and additional credits, demonstrates a commitment to customer satisfaction despite initial miscommunication.
The transparency shown by Thariq from the Claude Code team in addressing the bug and its resolution is a positive step towards improving user trust.

Concerns

The initial refusal to issue compensation for technical errors reflects poorly on Anthropic's customer service policies, raising concerns about their accountability.
Many users suspect that the automated responses are generated by an AI, highlighting potential issues with the company's reliance on AI for customer support.