df-infrastructure
Skill4 starsChanged 2 months agoApache-2.0View on GitHub
Dark Factory Infrastructure stage: produce the Deployment and Infrastructure Spec — DTAP placement, where data lives, trust boundaries mapped to enforcement mechanisms, secrets, and a verified observability surface. Triggers on "infrastructure", "deployment", "trust boundary", "where data lives", "deploy across environments".
In this repository43 other files it gives its agents.
- skills/vercel-react-best-practices/AGENTS.mdAGENTS.md
- agent-notepadSkill
- architecture-reviewerSkill
- backend-service-auditorSkill
- channel-listenerSkill
- code-reviewerSkill
- coder-file-transferSkill
- cold-start-generatorSkill
- creating-skillsSkill
- critical-thinkingSkill
- dark-factory-buildSkill
- dependency-risk-auditorSkill
- df-adversary-gateSkill
- df-app-walkthroughSkill
- df-context-storeSkill
- df-data-transform-lensSkill
- df-dispatch-subagentsSkill
- df-observabilitySkill
- df-product-ownerSkill
- df-qaSkill
- df-solution-architectSkill
- df-tdd-developerSkill
- df-ui-verifySkill
- df-video-intakeSkill
- dx-maintainability-reviewerSkill
- frontend-auditorSkill
- handoffSkill
- microservices-architectSkill
- production-readiness-reviewerSkill
- repo-docsSkill
- requirements-discoverySkill
- sc-auditSkill
- sc-devSkill
- scope-initSkill
- scope-retireSkill
- security-threat-modelerSkill
- service-mapperSkill
- test-quality-gatekeeperSkill
- vercel-react-best-practicesSkill
- vinculum-loopSkill
- vinculum-mapSkill
- web-design-guidelinesSkill
- work-autonomouslySkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 24
Start from the platform's standard environment progression (Dev → Test → Acceptance → Production), base + overlays.
- Very specific: line 26
Verify the dashboards render live data and the $correlationId query returns real cross-service results.
- Distinctive: line 27
Write the deploy procedure QA follows to stand the system up.
- Distinctive: line 30
Every service has an overlay per target cluster.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Dark Factory — Infrastructure Architect (where data lives + the boundaries)
Overview
Infra decides where the architecture runs and makes it deployable across all environments. It produces one must-have output: the Deployment & Infrastructure Spec. Through the lens (df-data-transform-lens): Infra places the data nodes and draws the boundaries. Runs in parallel with the Developer lane.
When to use
Deploying across DTAP, provisioning clusters/secrets/network/storage, wiring observability sinks + operational knobs, or enforcing runtime trust.
What Infra does
locationper data node → storage class, region, residency. Thegovernancetag (PHI/PII, residency) is enforced by where you put the data.- Every unit edge is a trust boundary. "Inside the cluster is trusted" is the sticky-trust anti-pattern. Secrets/IAM enforce the boundary, not network position. Map every SA runtime-trust entry to a mechanism (IRSA/IAM, sealed-secrets, network policy, DB roles).
- Build + verify the Observability Surface — stand up the dashboards + datasources + log/trace sinks and verify they render live data (the acceptance bar; see
df-observability). Ship dashboards as code (deployments/grafana/…), not hand-clicked. - Provision every operational knob the Ops runbook + reconciliation + effect-compensation paths will need.
Reversibility (the autonomy gate)
Building manifests is reversible and high-autonomy; applying them to production is irreversible → a human gate regardless of how green the evidence is (see df-adversary-gate / the two-axis autonomy model). A production change requires a ticket + announcement.
Instructions
- Start from the platform's standard environment progression (Dev → Test → Acceptance → Production), base + overlays. Record per-product deltas only.
- Map every SA runtime-trust entry → enforcement; every observability entry → a sink; every knob → a provisioning step.
- Verify the dashboards render live data and the
$correlationIdquery returns real cross-service results. - Write the deploy procedure QA follows to stand the system up.
Exit gate
Every service has an overlay per target cluster; every secret a source + rotation; every alert a sink; every knob a provisioning step; no implicit trust; ≥3 replicas on data-path services; the Observability Surface verified rendering live data. Independent of the Developer lane.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

