Read git history for the human risks — files only one person understands, critical paths with a truck factor of one, and the knowledge that leaves when they do.
Free & open · no signup · read-only — it ends by asking · nothing leaves your machine
You are working inside this repo. Mission: read the history for human risk — the files only one person has ever touched, the critical paths whose every author has left, and the knowledge that walks out the door when one person does.
Read-only pass. Lean on git log, git blame, and git shortlog. Your only write is the report file.
Phase 2 sweeps the codebase through every one of these, citing file and line for each finding.
areas where one author owns nearly all the knowledge; the count of people whose loss would strand a critical path
files whose every significant author has gone quiet or left; owned by nobody now
the money/auth/data paths cross-referenced with ownership: single points of human failure
high-churn or high-complexity files (see 22) that also have exactly one author: hardest to hand off
is there an ownership file, and does it match who actually commits, or is it aspirational
do the single-owner areas have any written explanation, or does the knowledge live only in one head
the areas a new hire could never safely touch without the one person who understands them
One structured report at the repo root — or in reports/, if you keep one — the same shape every time, ready for a teammate — or the optional Studio — to act on.
Every Goal Prompt follows the same four steps, so results are consistent and repeatable — no matter which one you run.
Build a picture from history.
Name files and contributors (by handle) for every finding.
Separate genuine risk (a critical path with a truck factor of one) from the harmless (a solo-owned throwaway script).
Create OWNERSHIP.md at repo root.
Copy it and paste it into your agent inside the repo you want checked.
Install the goal plugin once — two commands — then just type /goal:ownership-bus-factor.
/plugin marketplace add GhostlyGawd/goal-prompts/plugin install goal@goal-promptsOr install only this Goal Prompt as /goal-ownership-bus-factor:
curl -fsSL https://goal-prompts.vercel.app/install | BRIEF=72 shLet an agent fetch it mid-conversation, or pull the raw Goal Prompt by URL.
https://goal-prompts.vercel.app/raw/72.mdNothing hidden — this is the whole Goal Prompt, verbatim. Read it in a minute, edit it, or copy it as-is.
# Goal: Ownership & Bus Factor You are working inside this repo. Mission: read the history for human risk — the files only one person has ever touched, the critical paths whose every author has left, and the knowledge that walks out the door when one person does. Read-only pass. Lean on `git log`, `git blame`, and `git shortlog`. Your only write is the report file. ## Phase 1 — Reconstruct who knows what - Build a picture from history: per area, who authored it, who last touched it, how concentrated the authorship is. - Cross-reference with the blast zones — auth, payments, data, the build — from earlier audits or a quick read. - Note the signals of departure: top authors with no commits in the last N months. ## Phase 2 — Audit through 7 lenses Name files and contributors (by handle) for every finding. 1. **Truck factor** — areas where one author owns nearly all the knowledge; the count of people whose loss would strand a critical path 2. **Orphaned code** — files whose every significant author has gone quiet or left; owned by nobody now 3. **Critical-path concentration** — the money/auth/data paths cross-referenced with ownership: single points of human failure 4. **Silent complexity** — high-churn or high-complexity files (see 22) that also have exactly one author: hardest to hand off 5. **CODEOWNERS reality** — is there an ownership file, and does it match who actually commits, or is it aspirational 6. **Documentation cover** — do the single-owner areas have any written explanation, or does the knowledge live only in one head 7. **Onboarding barrier** — the areas a new hire could never safely touch without the one person who understands them ## Phase 3 — Curate - Separate genuine risk (a critical path with a truck factor of one) from the harmless (a solo-owned throwaway script). - For each risk: what breaks if that person is unavailable, and the cheapest way to spread the knowledge (pairing, docs, a guided change). - Rank by criticality × concentration. ## Phase 4 — Report Create `OWNERSHIP.md` at repo root: 1. **Bus-factor verdict** — the lowest truck factor on a critical path, named 2. **Risk map** — area · primary owner · # who understand it · criticality · status 3. **Orphans** — code no active contributor owns 4. **De-risk plan** — the three handoffs worth doing first, and how Start the report with today's date. If `OWNERSHIP.md` already exists from a previous run, read it first and lead with what changed since. ## Rules - History is evidence; name the files and the commit pattern behind each claim - Describe roles and risk, not judgments of people - No multi-contributor history in this repo? Say so in a one-paragraph null report and stop — a null result is a valid finding. - If a `reports/` directory exists at the repo root, write the report there instead of the root. - Before asking, present the top findings as a ranked list in plain words - Report only — end by asking which handoffs to start
reports/ directory exists at the repo root, write the report there instead of the root.Duplication, tangles, and dead patterns — mapped as interest paid vs principal owed, with a refactor sequence that pays for itself.
Walk clone → install → run → change → PR like a brand-new developer and log every stumble on the way to a first contribution.
Census the drifted colors, spacing, and duplicate components; propose the token set and migration order to unify them.
Can a coding agent work here? Entry files, one-command checks, discoverable conventions, and a mechanical definition of done — audited by attempting the work.
Runs inside these playbooks — curated sequences you can launch with one paste: