abahocodes/LLMGraph Invoke
Invoke a deployed LLMGraph workflow from CI and capture its JSON response.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Activelast commit Jul 15, 2026
- License
- MIT
Pinned Snippet
uses: abahocodes/llmgraph-invoke-action@1d10a457dd2c81c85afd7b15b9852c1bc5bbe1da # v1.0.0tags can be moved; commit SHAs can't. why a SHA?
Inputs
| name | description | required | default |
|---|---|---|---|
| endpoint | Full deployment endpoint URL from the LLMGraph dashboard, for example https://llmgraph.ai/api/<graph_id>/<environment>. | yes | — |
| api-key | LLMGraph deployment API key. Always pass this via a secret, for example ${{ secrets.LLMGRAPH_API_KEY }}. Never hardcode it. | yes | — |
| body | JSON object string to send as the request body. | no | {} |
| user-input | Convenience string merged into the body as the user_input field, the conventional input key for LLMGraph workflows. | no | "" |
| timeout-seconds | Per-request timeout in seconds. | no | 120 |
| fail-on-workflow-error | When true, a failed workflow run (HTTP 422) or a timed out run (HTTP 504) fails the step. When false, the step succeeds and the status output carries the outcome. | no | true |
Outputs
| name | description |
|---|---|
| response | Raw JSON response body returned by the deployment. |
| status | HTTP status code of the final response. |