Cost Estimation (--cost-estimate)

--cost-estimate adds a directional planning estimate to flaglint audit output. It is a planning guide — not a savings calculation, ROI projection, billing analysis, or LaunchDarkly contract cost estimate.

Usage

npx flaglint audit ./src --cost-estimate

npx flaglint audit ./src --cost-estimate --hourly-rate 125

Example Output

✓ Audit complete: 13 flags — 3 high risk, 10 medium risk

Migration readiness: 50/100  ·  moderate
[█████████████░░░░░░░░░░░░] 50%
10 safely automatable  ·  10 require manual review

Estimated migration effort: 22.8h – 43.9h
Estimates are directional. See the report for assumptions.

Algorithm

The estimate is derived from the number of automatable and manual-review call sites detected in the scanned directory.

Phase: automation

automationLow  = automatableCalls × 0.25
automationHigh = automationLow × 1.5

Phase: manual review

manualLow  = manualReviewCalls × 1.5
manualHigh = manualLow × 2

Phase: validation

validationLow  = (automationLow + manualLow) × 0.3
validationHigh = (automationHigh + manualHigh) × 0.3

Totals

Totals are computed from unrounded phase values, then rounded to one decimal place, then floored to a minimum of 4 hours for very small codebases:

totalLow  = max(4, round1(automationLow + manualLow + validationLow))
totalHigh = max(4, round1(automationHigh + manualHigh + validationHigh))

Default assumptions

Parameter	Default	Meaning
`automationHoursPerCall`	0.25	Hours per automatable call site
`manualReviewHoursPerCall`	1.5	Hours per manual-review call site
`validationMultiplier`	0.3	Validation overhead as a fraction of automation + manual total
`minimumHours`	4	Floor applied to both low and high totals

These defaults are planning heuristics, not observed industry benchmarks.

Hourly Rate (—hourly-rate)

Supplying --hourly-rate <number> adds costLow and costHigh fields to the top-level estimate object in JSON output. These fields are also included in markdown and HTML report output. --hourly-rate is valid only when --cost-estimate is also specified.

npx flaglint audit ./src --cost-estimate --hourly-rate 125

Not-Applicable Behavior

When no direct LaunchDarkly calls are detected, the estimate is null and no estimate line is printed. This matches the not-applicable readiness score state.

Disclaimer

Estimates are directional planning guides based on call-site complexity. Actual effort depends on test coverage, team familiarity, and provider setup. FlagLint does not access runtime data or LaunchDarkly billing.