ganakailabs/Cloudeval Actions

Run CloudEval CLI in GitHub Actions with scoped access keys, gating, reports, PR feedback, and artifacts.

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
access_keyCloudEval access key (repository secret; use secrets.CLOUDEVAL_ACCESS_KEY).yes
base_urlOptional API base URL (CLOUDEVAL_BASE_URL / cloudeval --base-url).no""
modeExecution mode — review, ask, gate, agent, reports, or nightly.noask
project_idProject id for project-scoped commands.no""
ask_promptPrompt for ask (or gate/nightly when using ask).no""
agent_taskTask for agent mode (when set, ask/gate/nightly use agent instead of ask where applicable).no""
gate_thresholdNumeric threshold for gate mode (optional on ask/agent/nightly to enforce a bound).no""
gate_jqjq expression on CLI JSON stdout yielding one number (tonumber applied).no.score
gate_operatorHow extracted value relates to threshold — ge (>=), gt, le, lte (<=), lt, eq, ne.noge
summary_answer_jqOptional jq expression to pull a short text snippet into the job/PR summary (e.g. .answer).no""
quietPass --quiet to cloudeval for cleaner CI logs.notrue
progressCLI progress — use none to pass --progress none (recommended in CI).nonone
modelOptional cloudeval --model.no""
profileOptional cloudeval --profile.no""
reports_typecloudeval reports run / download --type.noall
reports_regioncloudeval reports run --region (cost reports).noeastus
reports_currencycloudeval reports run --currency.noUSD
reports_waitPass --wait to reports run (poll until jobs finish).nofalse
reports_poll_interval_msPoll interval when reports_wait is true.no2500
reports_downloadRun reports download after reports run.notrue
reports_output_dirOutput directory for reports download.nocloudeval-reports
repoGitHub repository for review mode. Defaults to github.repository.no""
refGit branch/ref for review mode. Defaults to github.ref_name.no""
commit_shaCommit SHA for review mode. Defaults to github.sha.no""
source_rootOptional IaC source root for GitHub-backed CloudEval projects.no""
config_pathOptional path to .cloudeval/config.yaml for review gates.no""
ignore_dirtyPass --ignore-dirty to cloudeval review.nofalse
review_waitWait for GitHub sync and report refresh before writing review output.notrue
review_wait_timeout_msMaximum time cloudeval review waits for GitHub sync.no900000
review_poll_interval_msPoll interval for cloudeval review while waiting.no5000
ai_summaryInclude an AI-written summary in review.json, review.md, and PR comments.notrue
ai_summary_modeAI summary mode for review output — ask or agent.noask
ai_summary_profileAgent Profile id when ai_summary_mode is agent.noarchitecture
review_output_dirOutput directory for review.json and review.md.nocloudeval-review
working_directoryWorking directory for CLI.no.
cli_install_urlCloudEval CLI install script URL.nohttps://cli.cloudeval.ai/install.sh
skip_cli_installSkip installer; cloudeval must be on PATH (mirrors, tests).nofalse
post_pr_commentPost or update one PR comment (pull-requests write; limited on fork PRs).nofalse
pr_comment_collapsed_detailsWrap the summary body in a collapsible <details> block in the PR comment.nofalse
pr_comment_json_excerptAppend a collapsible JSON excerpt from CLI stdout (truncated) to the PR comment.nofalse
pr_comment_max_json_charsMax characters of JSON excerpt in PR comment.no12000
upload_artifactsUpload staged output (summary, json, reports) as a workflow artifact.nofalse
artifact_nameArtifact name when upload_artifacts is true.nocloudeval-output
artifact_retention_daysArtifact retention (days); 0 uses repo default.no14
include_run_metadataAdd workflow metadata to markdown summary and PR comment. Review mode preserves CLI workflow/artifact badges when available and appends metadata only for older summaries.notrue
job_summary_titleHeading for the GitHub Actions job summary.noCloudEval
namedescription
resultpass or fail
scoreNumeric value when gating ran (same as extracted_value)
extracted_valueNumeric value from gate_jq when gating ran
summary_markdownMarkdown summary
summary_filePath to summary file on the runner
report_pathAbsolute path to reports directory when download ran
json_pathPath to captured CLI JSON stdout
artifact_pathPath to staged artifact directory before upload
run_urlURL to this workflow run (when include_run_metadata supplies server env)