--- # Machine-readable header. Unknown = null. Uncertain values: add comment "# assumed". # Missing owner input: add comment "# OPEN: ask owner". handover_version: 0.3 title: "" # short project name summary: "" # one sentence: what is it? kind: null # web-app | desktop-app | mobile-app | cli-tool | service-bot | library | game | game-map # | 3d-model | firmware | hardware | asset-pack | prototype | other language: null # language of code comments / UI texts, e.g. de, en tags: [] # free-form, e.g. [pcb, svelte, computer-vision] stack: [] # languages, frameworks, engines, tools completion_percent: null # measured against the project goal, see section 2 remaining_effort_hours: null # hands-on hours of an experienced developer until "usable" last_activity: null # YYYY-MM-DD, last real work invested: { hours: null, money_eur: null } # human hours + real expenses, NOT AI runtime ai_generated_share: null # none | some | most | all license: null # SPDX identifier, e.g. MIT, CC-BY-4.0 third_party_rights: null # none | listed | unclear (see section 9) secrets_removed: null # true only after the owner has cleaned up publish_check: null # BLOCK | WARN | OK (see section 9) offer: { type: null, price_eur: null } # free | fixed | wish | auction source: null # e.g. https://codeberg.org/name/repo, a download link, or "upload" runs_without_external_services: null # true/false; what is missing then? -> section 4 --- # ## 1. What is it? What does the project do, which problem does it solve, for whom? (3–6 sentences) ## 2. Status: what works, what doesn't | Part | Status | Note | |---|---|---| | … | done / partial / broken / not started | … | How was this verified (tests, manually, not at all)? Include the date. ## 3. Structure Folder layout and key files, one sentence each. How do the parts fit together? ## 4. Getting it running Exact commands from "fresh clone" to "running". Required accounts, keys (name only, never the value), hardware. ## 5. Decisions and abandoned approaches What was done deliberately, and what was tried and dropped – with the reason. This saves the next person the most time. ## 6. Why did you stop? Be honest: no time, technically stuck, lost interest, no market, too expensive … Would that obstacle also stop someone else? ## 7. Next steps The 3–5 most sensible next steps, in order, each with effort and the file to start in. ## 8. Known pitfalls Things you only learn the hard way: platform quirks, bugs, environments, stubs/dead code, cache or version switches, unsecured services that cost money. ## 9. Rights and content - **Own work:** what you made yourself - **AI share:** which parts are AI-generated, with which tool - **Third-party:** libraries, assets, models, photos – with license - **Redistribution:** may everything be passed on? - **Secret scan:** table `Finding | Location | Assessment (real / test data / false positive)` – type and location only, never values - **Publish check:** `BLOCK` | `WARN` | `OK` – one line with the reason ## 10. For agents Where should an agent start reading? Which files are the source of truth, which are outdated (name the concrete contradiction)? How does it verify its work (test command)? ## Open owner questions Numbered list of all `OPEN` items. Empty once everything is answered.