← ClaudeAtlas

hindsightlisted

Deep retrospective architectural review of an entire project against a user-stated goal, producing ranked, actionable findings in inHindsight.md. Use this skill whenever the user runs "/hindsight", asks for an "architectural review", "codebase review", "technical debt review", or "hindsight review", asks "what should I improve here", "what would you have done differently", or "review this project against my goal", or wants to re-check which previously reported findings have been resolved. Trigger even if the user doesn't name the skill but describes reviewing a whole project (not a single file) against a purpose or goal. Requires the user to state the project's goal; writes the review to inHindsight.md (plus a machine-readable inHindsight.json) at the project root. Also runs unattended — for CI, headless runs, or another agent delegating a review.
tansharma/hindsight · ★ 0 · Data & Documents · score 59
Install: claude install-skill tansharma/hindsight
# Hindsight — Retrospective Architectural Review Hindsight reviews an entire project the way a Senior Technical Architect reviews a system that has already been built: with full knowledge of how it turned out. It is for experienced (mid-to-senior) engineers, so findings should be substantive — not linting, not style nits, not isolated-file observations. Every run produces one deliverable: `inHindsight.md` at the project root, containing findings ranked by Impact/Effort, each with a ready-to-paste LLM prompt that would fix it. ## The Hindsight Principle The skill is named for a specific kind of judgment: understanding that is only available *after* the fact. Decisions that were reasonable — even obviously correct — when they were made can turn out to be wrong once the whole project exists and its real purpose is known. That is the judgment this review must apply, and it is what separates Hindsight from a linter or a per-file code review. In practice this means the review happens in two distinct passes: 1. **Reconstruct what was intended.** Before judging anything, learn what the authors believed and planned *at the time*: READMEs, design docs, ADRs, spec files, TODO/FIXME comments, and — when the project is a git repository — the commit history (`git log --oneline` shows the order in which decisions solidified). Record intent as written, before reading the code that resulted from it. 2. **Judge the outcome against that intent and the user's goal.** Now read