Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
54 changes: 14 additions & 40 deletions .github/workflows/skill-evals.yml
Original file line number Diff line number Diff line change
Expand Up @@ -4,11 +4,9 @@
# See LICENSE for license information.
###############################################################################

# Grades this repo's skills with https://github.com/danielholanda/skillscope:
# structure (free), then routing (which skill fires) and behavioral (what the
# agent did once it fired).
# Run skillscope on evals.json (same intake criteria as amd/skills)

name: skill-evals
name: AMD Skills Checks

on:
pull_request:
Expand All @@ -21,39 +19,15 @@ permissions:
contents: read

jobs:
structural:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- uses: danielholanda/skillscope@main
with:
command: structural
skills: TraceLens/Agent/Analysis/skills/analysis-orchestrator

routing:
needs: structural
runs-on: ubuntu-latest
timeout-minutes: 40
env:
ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
steps:
- uses: actions/checkout@v4
- uses: danielholanda/skillscope@main
with:
command: routing
skills: TraceLens/Agent/Analysis/skills/analysis-orchestrator
install-claude: "true"

behavioral:
needs: structural
runs-on: ubuntu-latest
timeout-minutes: 90
env:
ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
steps:
- uses: actions/checkout@v4
- uses: danielholanda/skillscope@main
with:
command: behavioral
skills: TraceLens/Agent/Analysis/skills/analysis-orchestrator
install-claude: "true"
evals:
name: AMD Skills Checks
uses: danielholanda/skillscope/.github/workflows/reusable.yml@main
secrets:
api_key: ${{ secrets.ANTHROPIC_API_KEY }}
with:
skills: TraceLens/Agent/Analysis/skills/analysis-orchestrator
# Each step can be set to `required`, `optional`, or `off`.
# On amd/skills, all steps are required.
structural: required
routing: required
behavioral: required
Loading