Playbook
Ship MVP
The work this situation takes, including the parts that are easy to miss. Start with any one.
Project scoring and MVP pick
A scored, ranked project pipeline with a recommended MVP that clears your hard gate.
Slugify function
A correct slugify(text) that passes the held-out cases.
Deploy pipeline and hosting
The app is deployed and the health endpoint returns 200.
Security baseline (dependency + CVE scan)
A dependency/CVE scan runs and reports zero unresolved criticals.
Model router that sends each query to the right LLM
A routing function that dispatches each query to the right model by its type, normalizing case and falling back to a default for any unlisted type.
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.
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.
Product requirements document (PRD)
A complete PRD: the named sections, a quantified business context, a measurable success metric, and acceptance criteria with a clear out-of-scope fence.
User story with acceptance criteria
A complete user story in the three-clause form with given/when/then acceptance criteria, a quantified reach, and an out-of-scope boundary.
Three-point (PERT) effort estimates
A PERT estimate per task (expected, margin, low, high) and a rolled-up total, from your optimistic/most-likely/pessimistic inputs.
Requirements traceability matrix
A matrix linking every requirement to its tests (and back), with computed orphans and a coverage percentage that clears your bar.
OpenAPI endpoint reference
A complete, well-formed reference for one endpoint: method, path, parameters (with every path placeholder covered), the required status codes, and a consistent request body.
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.
Test coverage with a release gate
Statement/branch/function coverage percentages from executed-vs-total counts, an overall aggregate, and a pass/fail gate against your per-type minimums.
Maintainability index per module
The Maintainability Index (raw and 0-100 normalized) per module from Halstead volume, cyclomatic complexity, and lines of code, with a maintainability band.
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.
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.
Compute Amdahl's-Law parallel speedup and efficiency
A pure function that, given a parallelizable fraction and processor count, returns Amdahl's-Law speedup, efficiency, and the asymptotic ceiling, graded on hidden numeric cases.
Also part of this playbook, opening soon:
- Add secure sign-in tokens
- Help people find my site on search
- See how people use my product
- Get alerted when my app breaks
- Write the launch announcement