Free tools for software-delivery teams

Maps 5 dimensions of a story (complexity, scope, unknowns, dependencies, testing effort) to a Fibonacci estimate (1, 2, 3, 5, 8, or 13). Or flags it for splitting.

Framework: Teaches the 5-dimension framework: estimating effort directly collapses risk into a fake-precise number; estimating across independent axes produces calibrated results.

Calibrated against ~3,000 stories from Stride beta customers.

Takes 6 inputs (team size, sprint length, PTO, meetings, on-call, velocity) and outputs realistic sprint capacity in points, with a breakdown of where time goes.

Framework: Naive capacity (team × days) overestimates by 40-50%. PTO + meetings + on-call rotation are the three biggest gaps; this calculator surfaces each so teams stop systematically over-committing.

Calibrated against ~400 sprint plans from Stride beta customers.

Estimates code-review wall-clock time + quality risk from a PR's size, files changed, tests, and cross-module scope. Outputs an easy / medium / hard / split tier.

Framework: PR size predicts review quality. Cohen 2006 + Bosu et al. 2017 + GitHub 2023 data converge: review quality drops sharply past 400 LOC, becomes essentially impossible past 800. The calculator quantifies it so teams can self-regulate.

Grounded in published research (Cohen 2006, Bosu 2017, GitHub 2023).

Takes 4 inputs (team size, hourly cost, hours saved per week, monthly seat fee) and outputs annual savings, cost, ROI multiple, and payback days. Conservative defaults.

Framework: Hours saved × hourly cost minus seat fees. Marketing-leaning, we say so on the page, but every input is yours to control. Defaults reflect mid-market US engineering economics with conservative hours-saved (5/week, the low end of Stride beta telemetry).

Hours-saved range from telemetry across 5,200 stories on Stride beta customers.

Enter the four DORA inputs (deploys/month, lead time, MTTR, change failure rate) and get composite Elite / High / Medium / Low tier with per-dimension explanation.

Framework: DORA's honest framing: your composite tier is bounded by your weakest dimension, not the average. Calibrated against Google's 2024 State of DevOps thresholds; each tier comes with a one-sentence explanation citing the underlying threshold.

Thresholds from Google's 2024 State of DevOps report.

Fill in context, decision, alternatives considered, and consequences. Get a Michael-Nygard-style ADR in markdown that you can copy or download.

Framework: The 4-section ADR (context / decision / alternatives / consequences) is the canonical format for capturing architecture decisions. The generator enforces the structure so the rationale survives team changes.

Format from Michael Nygard, 2011. Used by thousands of engineering teams.