Add EvoSkill-based agent skill evolution workflow - #48722
Merged
Conversation
Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>
Copilot
AI
changed the title
Add EvoSkill-based skill evolution workflow
Add EvoSkill-based agent skill evolution workflow
Jul 28, 2026
Copilot created this pull request from a session on behalf of
pelikhan
July 28, 2026 20:20
View session
pelikhan
marked this pull request as ready for review
July 28, 2026 20:20
Contributor
There was a problem hiding this comment.
Pull request overview
Adds a scheduled EvoSkill workflow for evolving repository skills from agent-run failures.
Changes:
- Adds role-separated skill evolution and held-out validation.
- Restricts mutations to five files under
.github/skills/**. - Adds the compiled workflow.
Show a summary per file
| File | Description |
|---|---|
.github/workflows/evoskill-evolver.md |
Defines the evolution workflow. |
.github/workflows/evoskill-evolver.lock.yml |
Provides the compiled GitHub Actions workflow. |
Review details
Tip
Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Comments suppressed due to low confidence (1)
.github/workflows/evoskill-evolver.md:190
- This validator contract also assumes an original skill, so it is undefined for
operation: create. Tell the validator how to score an absent baseline to keep the supported creation path usable.
Evaluate the original skill and candidate diff independently against only the supplied held-out samples. Penalize over-broad triggers, memorized examples, unverifiable procedures, and regressions on successful cases. Do not edit files.
- Files reviewed: 1/2 changed files
- Comments generated: 2
- Review effort level: Medium
|
|
||
| ### 1. Build stratified train and validation sets | ||
|
|
||
| Use the GitHub Actions tools to inspect completed agentic workflow runs from the last 14 days. Exclude this workflow and ordinary non-agentic workflows. |
|
|
||
| ### 5. Score on held-out validation | ||
|
|
||
| Only now give the `validator` agent the held-out validation set, the original target skill, and the candidate diff. The validator must score both baseline and candidate independently from 0–100 using: |
Contributor
|
🎉 This pull request is included in a new release. Release: |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Introduces an agentic workflow based on EvoSkill’s failure-driven skill discovery method. It evolves repository skills while keeping models and non-skill code unchanged.
Evolution loop
Selection
Safety
.github/skills/**.