project-postmortemlisted

Generate an evidence-grounded project retrospective — pulled from git history and decision docs, not memory — that grades the project on separate dimensions, splits what went right from wrong, reduces the failure to its root ordering error, extracts transferable lessons for the next project, and names what survives. Use whenever a project ships, stalls, or gets sunset and someone wants the writeup — "post-mortem", "retrospective", "what went wrong with X", "we're sunsetting X, write it up", or mid-project "why did this go sideways". The method cites commit hashes or sources for every claim, builds a timeline of when validation happened vs building, separates pros from cons including uncomfortable symmetries, and ends with a per-dimension scorecard plus a what-survives list — then runs a check that fails any vanity scorecard or uncited narrative. Distinct from drift-check (audits live docs) and conductor-memory/session-bookend (capture session state for continuity); this extracts backward-looking lessons.
scoobydrew83/skills · ★ 0 · AI & Automation · score 78

Install: claude install-skill scoobydrew83/skills

# Project Post-mortem Write the retrospective a project earns — grounded in what the git history and decision docs actually show, graded honestly on separate axes, reduced to the one decision that mattered most, and mined for lessons that transfer to the *next* project. The one rule everything serves: > **Ground every claim in retrieved evidence, not memory. If you can't cite it, label it as recollection.** A post-mortem written from memory is a story the team tells itself, and stories drift toward flattering the teller. A post-mortem written from the commit log, the PRs, and the decision docs is a record — and a record can teach the next project something the memory would have smoothed over. --- ## Why this exists The value of a post-mortem is almost entirely in two things: an *honest* root cause, and *transferable* lessons. Both are easy to fake and easy to get wrong: - **Memory launders failure.** By the time a project ends, everyone has a tidy narrative. The git history doesn't — it shows that the validation commit came on day four, not day one, regardless of how it felt. Cite the evidence and the real story surfaces. - **A blended grade hides the killer.** "B+ overall" averages away the fact that engineering was an A and differentiation was an F — and the F is the whole lesson. Grade dimensions *separately* or you erase the finding. - **Vanity retros teach nothing.** A retro that's all "we learned a lot and grew as a team" is a feeling, not a lesson. The test of