Skip to content

docs: add hermes-eval — skill regression testing + trajectory scoring#27071

Closed
Saurav0989 wants to merge 1 commit into
NousResearch:mainfrom
Saurav0989:main
Closed

docs: add hermes-eval — skill regression testing + trajectory scoring#27071
Saurav0989 wants to merge 1 commit into
NousResearch:mainfrom
Saurav0989:main

Conversation

@Saurav0989

Copy link
Copy Markdown
Contributor

Adds hermes-eval — a skill regression testing + trajectory quality scoring library.

Addresses the skill drift problem documented in issue #13737 and multiple community posts.

The Atropos adapter outputs quality-scored ShareGPT trajectories directly in the format Atropos RL environments consume. MIT licensed, all tests offline (no LLM calls required).

@alt-glitch alt-glitch added type/docs Documentation improvements P3 Low — cosmetic, nice to have labels May 16, 2026
teknium1 added a commit that referenced this pull request May 17, 2026
@teknium1

Copy link
Copy Markdown
Contributor

Merged via PR #27247 — your commit was cherry-picked onto current main as part of a batch salvage of low-risk new-contributor PRs. Authorship preserved (docs: add hermes-eval to Community). Thanks for the contribution — appreciate it.

@teknium1 teknium1 closed this May 17, 2026
@teknium1

Copy link
Copy Markdown
Contributor

Heads up — I merged this earlier today as part of a batch salvage, but on a closer look at the linked repo (2 commits, 0 stars, 0 forks, no published releases) I've removed the link from the README in #27271. Nothing personal — README links from us implicitly endorse community projects, and we shouldn't be linking brand-new repos before they have any usage or track record. If hermes-eval picks up users and a real activity signal, please re-submit and we'll add it back.

gweeteve pushed a commit to gweeteve/hermes-agent that referenced this pull request Jun 2, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

P3 Low — cosmetic, nice to have type/docs Documentation improvements

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants