Choose what to get done
AI coding session brief and context file
The founder gets a structured session-brief template plus a populated project-context document so each AI coding session executes against fixed architectural guardrails and logs what it changed, preventing the compounding drift the book warns about.
Alert rule register (with runbook links)
A deduplicated alert rule set where every paging alert links a runbook, cutting alert fatigue and giving on-call an immediate response path.
Blameless post-incident review register
A root-cause register where each incident carries a why-chain, a stated root cause, and owned, dated corrective actions, so the same incident does not recur.
Clean-code smell scorecard for a method
A per-method clean-code smell scorecard: which thresholds each method exceeds (long method, too many arguments, deep nesting, and more), a 0-100 score, and a pass/warn/fail verdict.
Cyclomatic complexity with refactor flags
McCabe cyclomatic complexity per function from its control-flow graph, with a refactor flag for any that exceed your threshold.
Defect density per module with ranked hotspots
Defect density per module against your chosen size unit, the overall density, and a ranking of the most defect-dense modules.
Defect removal efficiency
Defect Removal Efficiency from pre-delivery vs escaped defect counts, with an optional per-phase breakdown.
Deploy runbook with rollback
An ordered, role-assigned deploy runbook with preflight checks, deploy steps, verification, and a rollback path that every failure step routes to.
Deployment pipeline and hosting
The app is deployed and the health endpoint returns 200.
Goal decomposed into an executable task plan
A goal broken into ordered, tool-assigned steps with a resolved, acyclic dependency graph, so an agent (or a team) can execute it without ambiguity.
Incident response runbook
A severity-tiered response playbook (roles, response targets, escalation, comms) so responders follow a plan instead of debating process mid-outage.
Input guard against prompt injection
A reusable guard that enforces type, length, and charset per field and rejects inputs smuggling section-break delimiters or fake instruction/role markers, before the LLM sees them.