SpectreHub discovers installed audit tools, runs them against your configured infrastructure, and produces one unified report with health scores, trends, and prioritized recommendations.
SpectreHub does not inspect infrastructure directly. It orchestrates specialized audit binaries and aggregates their deterministic output.
One command. All your infrastructure. Actionable results.
$ spectrehub run
Discovered 4 tool(s), 4 runnable
vaultspectre ✓ ready
pgspectre ✓ ready
kafkaspectre ✓ ready
clickspectre ✓ ready
Total Issues: 40 (↓ 9% from last week)
Health Score: GOOD (91%)
Recommendations:
1. [CRITICAL] Fix 8 missing Vault secrets
2. [HIGH] Clean up 7 unused Kafka topics
3. [MEDIUM] Review 18 stale ClickHouse tablesNot a vanity number.
Health score is a weighted deduction model. Each issue is associated with a resource. The score represents the percentage of resources without active issues, weighted by severity — critical deducts more than medium. Clamped to 0-100. No ML, no heuristics — deterministic arithmetic you can reproduce from the JSON output.
Three steps. Zero config locally, minimal config in CI.
SpectreHub checks which tools are installed and which targets are configured via environment variables and config files. Read-only — no secret values are ever printed or uploaded. SpectreHub never transmits infrastructure metadata without explicit configuration.
Each detected tool runs with --format json, capturing structured output. Per-tool timeouts and partial failure handling built in.
Reports are normalized into a unified issue format, scored, compared against previous runs, and turned into prioritized action items.
Every tool in the Spectre family is a standalone CLI binary that follows the ANCC convention.
Add infrastructure audits to any CI/CD pipeline in two lines.
- name: SpectreHub Audit
uses: ppiankov/spectrehub-action@v1
with:
threshold: 50The action installs tools, discovers targets from repository secrets, runs audits, and posts results as a PR comment. Exit code 1 if issues exceed your threshold.
Real drift from real infrastructure.
SpectreHub found:
8 Vault secrets referenced in code but missing from Vault
7 Kafka topics with zero consumers for 30+ days
18 ClickHouse tables untouched in 90 days (412 GB)
3 PostgreSQL roles with superuser but no login in 6 months
2 MongoDB collections with no indexes on query patterns
Estimated monthly waste: $840 (based on public pricing for idle resources)
Time to find manually: 2-3 days across 4 teamsWaste estimates use public cloud pricing for identified idle resources. Optional, transparent, and overridable in config.
Free to audit. Paid to stop repeating work.
go install github.com/ppiankov/spectrehub/cmd/spectrehub@latestbrew install ppiankov/tap/spectrehub