caleb-todd-commits/Hedge Security Diff

Shows how a pull request changes attack surface, trust boundaries, privilege, and data flow.

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
commandcheck, collect, reason, publish, init, verify, or prunenocheck
openai-api-keyOpenAI API key. Prefer workload identity federation for production.no
github-tokenGitHub token used to post or update the PR report.no
rootRepository root to analyze.no.
config-pathPath to .hedge.yml.no.hedge.yml
model-triageOptional trusted workflow override for the low-cost triage model.no
model-analysisOptional trusted workflow override for the deep security reasoning model.no
base-refExact base Git revision for secretless offline collection. Must be paired with head-ref.no
head-refExact head Git revision for secretless offline collection. Must be paired with base-ref.no
offlineDisable all OpenAI requests even when a key is present.nofalse
collection-pathPath to the validated collection bundle for staged execution.no
collection-manifest-pathPath to the immutable collector RunManifest.no
reason-bundle-pathPath to the validated reason bundle for staged execution.no
run-manifest-pathInput path to the current stage RunManifest.no
workflow-refExact GitHub workflow ref recorded by the trusted caller.no
workflow-shaExact commit SHA containing the trusted workflow.no
action-refPublished Hedge Action reference pinned to a full commit SHA.no
risk-idExisting HEDGE-NNN identifier for verify or prune commands.no
verification-evidenceRepository-relative verification JSON path for the verify command.no
acceptance-reasonDocumented risk-acceptance reason for the prune command.no
acceptance-reason-b64Base64-encoded acceptance reason. Preferred in workflows to avoid multiline output injection.no
actorRecorded human or automation actor for lifecycle updates.no
dry-runGenerate output without posting comments or failing the check.nofalse
namedescription
surface-changedWhether an evidence-linked security architecture delta was detected.
open-risksNumber of open findings in this run.
highest-severityHighest finding severity.
decisionRecorded merge decision: allow, warn, or block.
report-pathPath to the generated Markdown report.
html-report-pathPath to the standalone interactive HTML dashboard.
sarif-pathPath to the generated SARIF 2.1.0 report.
delta-pathPath to the machine-readable security architecture delta.
analysis-pathPath to the complete machine-readable analysis result.
coverage-statusEvidence coverage: complete, partial, or unsupported.
analysis-statusAnalysis health: complete, degraded, or failed.
confirmed-no-deltaTrue only for exact revisions with complete supported coverage and no graph delta.
model-routeCost-aware model route: none, deterministic, triage, analysis, triage-analysis, or fallback.
model-callsNumber of completed OpenAI model calls reported for this run.
input-tokensProvider-reported input tokens across completed model calls.
output-tokensProvider-reported output tokens, including reasoning tokens, across completed model calls.
total-tokensProvider-reported total tokens across completed model calls.
run-manifest-pathPath to the integrity-bound RunManifest v0.1 artifact.
collection-pathPath to the secretless exact-revision collection bundle.
collection-manifest-pathPath to the immutable collector RunManifest.
reason-bundle-pathPath to the schema-validated reasoning bundle.