Issue Station
Sign In
Tools hereToolbox

hermes - 💡(How to fix) Fix [Bug]: EXCESSIVE BRANDING DETECTED BY ANTHROPIC! [2 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
NousResearch/hermes-agent#19046•Fetched 2026-05-03 04:52:41
View on GitHub
Comments
2
Participants
2
Timeline
5
Reactions
0
Author
Many0therFunctionsMany0therFunctions
Participants
alt-glitchalt-glitchMany0therFunctionsMany0therFunctions
Timeline (top)
labeled ×3commented ×2

Error Message

Additional Logs / Traceback (optional)

Root Cause

Root Cause Analysis (optional)

Code Example

Read above

---
RAW_BUFFERClick to expand / collapse

Bug Description

Is there a reason why you're practically screaming to anthropic's API that this is a third party harness???

<img width="1244" height="432" alt="Image" src="https://github.com/user-attachments/assets/8e02e219-2b25-4d1f-9211-0ebcf3d92ad2" /> <img width="1073" height="387" alt="Image" src="https://github.com/user-attachments/assets/4fc223fe-c23d-4eb5-b3ae-e0319ec39912" /> <img width="1301" height="367" alt="Image" src="https://github.com/user-attachments/assets/b31fb7a6-b051-4f55-84cb-8fa188a26ff2" />

I know other harnesses are still able to use anthropic's shared / subscription limits, but man you guys are just screaming to anthropic with these. I am shocked and appalled.

That's both wasting tokens AND being flagged for detection. Feels very... controlled opposition-ey....

Steps to Reproduce

  1. Run hermes and add anthropic model importing creds from claude code or freecode
  2. Do any task.
  3. Open up the claude usage tracker and notice it using your extra usage instead of the subscrtiption limits
<img width="986" height="173" alt="Image" src="https://github.com/user-attachments/assets/9c21d9ad-408c-4ea4-9db1-fb05c3b1bc5d" />

Expected Behavior

Convincingly passed off as an old version of claude code, and usage should have used up subscription limits first

<img width="983" height="164" alt="Image" src="https://github.com/user-attachments/assets/eef2d372-7106-40f7-931c-1908d06dfaa3" />

it should have been efficient with tokens and not insert nous branding into everything

Actual Behavior

It has intentionally retarded opsec and inserts hermes branding and mentions hermes in the prompt, which people have found out, gets flagged by anthropic as a third party harness. the Goal is to evade as much as possible. The goal is supposed to be a game of cat and mouse with these providers!

Affected Component

Agent Core (conversation loop, context compression, memory)

Messaging Platform (if gateway-related)

No response

Debug Report

Read above

Operating System

Debian

Python Version

3.11

Hermes Version

Hermes Agent v0.12.0 (2026.4.30)

Additional Logs / Traceback (optional)

Root Cause Analysis (optional)

No response

Proposed Fix (optional)

No response

Are you willing to submit a PR for this?

  • I'd like to fix this myself and submit a PR

extent analysis

TL;DR

Modify the API requests to convincingly pass off as an old version of claude code to avoid being flagged as a third-party harness.

Guidance

  • Review the code that constructs API requests to anthropic's API and modify it to remove any identifying information that could be used to detect hermes branding.
  • Verify that the usage tracker is using subscription limits first by monitoring the claude usage tracker after making the changes.
  • Investigate the conversation loop, context compression, and memory components of the Agent Core to ensure they are not inserting hermes branding or mentions into the prompts.
  • Test the changes in a controlled environment to ensure they do not introduce any new issues.

Example

No code snippet can be provided without more information about the specific API requests being made.

Notes

The goal of evading detection as a third-party harness may be subject to the terms of service of anthropic's API, and any changes made should ensure compliance with these terms.

Recommendation

Apply workaround: Modify the API requests to remove identifying information and test the changes in a controlled environment, as the issue seems to be related to the way hermes is interacting with anthropic's API.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#api#tensor shape#autograd error#model save/load#optimization

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING