Action Reference

Action Reference

→ Full docs: https://synapsekit.github.io/synapsekit-docs/docs/evalci/action-reference

---

Inputs

Input	Default	Description
path	.	Path to eval files or directory
threshold	0.7	Global minimum score (0.0–1.0)
extras	openai	pip extras e.g. openai,anthropic
synapsekit-version	latest	Pin a version e.g. 1.5.2
github-token	${{ github.token }}	For posting PR comments
fail-on-regression	false	Fail if scores regress vs baseline
token	(empty)	Reserved for future backend token

---

Outputs

Output	Description
passed	Number of cases that passed
failed	Number of cases that failed
total	Total cases run
mean-score	Mean score (4 decimal places)

- uses: SynapseKit/evalci@v1
  id: eval
  with:
    path: tests/evals

- run: echo "${{ steps.eval.outputs.passed }}/${{ steps.eval.outputs.total }} passed"

---

Exit codes

Code	Meaning
0	All cases passed
1	One or more cases failed

---

Provider extras

Provider	extras
OpenAI	openai
Anthropic	anthropic
Gemini	google-generativeai
Groq	groq
Multiple	openai,anthropic