← ClaudeAtlas

assumptionslisted

Generate an evidence-backed Assumptions ledger for a Git diff, feature, endpoint, worker, migration, or code path. Use when reviewing a change, asking what could break, preparing a release, finding edge cases, assessing deployment safety, analyzing retries or concurrency, or designing failure-focused tests.
Teycir/Assumptions · ★ 20 · Code & Development · score 71
Install: claude install-skill Teycir/Assumptions
# Assumptions ## Purpose Identify the conditions a change silently depends on in order to work safely in production. Produce a structured ledger of: - hidden assumptions, phrased as conditions that must hold - repository evidence, with a file and line locator where available - failure modes - existing safeguards - falsification tests - recommended controls - status, evidence confidence, and priority This is not a generic code review. Do not report vague concerns. ## Core standard Only report an assumption when you can provide all of the following: 1. The assumption stated as a falsifiable condition (what must hold, not what the current code does — see "Phrasing assumptions as conditions" below). 2. Direct evidence from code, tests, configuration, schema, documentation, or the Git diff, with a file path and line range when the tooling available can produce one. If a line range is unavailable, cite the file path and the symbol (function, route, migration file) instead. Never present a claim as repository evidence without a locator. 3. A concrete consequence if the assumption is false. 4. Existing safeguards, or an explicit statement that none were found. A "none found" statement must name the scope that was searched (files, directories, or symbols) — see "Search boundary for absence claims" below. Do not write "none found" without saying where you looked. 5. A practical falsification test or verification step. 6. A status label: Protected, Part