Story Done

Before you start: read this whole install prompt, tell me in a few lines what this skill does and what installing it involves, and wait for my go-ahead before you do anything else. Prerequisites — INTEGRATED SYSTEM FIRST: this skill is one piece of the complete "Claude Code Game Studios" template (49 agents, 74 skills, studio templates and a studio CLAUDE.md, per the repository README). Each skill assumes that template is installed, with its conventions, its project.yaml / project.local.yaml configuration, its .claude/docs/ mode documents, its hooks (for example hooks/yaml-helper.sh, outside this skill's folder) and its .claude/scripts/ helpers. Installing this listing alone copies only this skill's folder, without the rest of the system: the skill may reference template agents, files, hooks or scripts that are absent outside the full template, and it may not run as designed on its own. No account, no key and no payment are needed to copy the files, and nothing is run during installation. One disclosed exception to the usual ask-before-write pattern: its session checkpoint in production/session-state/active.md is written on every run, whatever the verdict; the story close itself still requires your explicit choice. Install "Story Done" for me. The skill itself closes a finished story — the only skill that marks a story Complete, and only on the user's yes. It finds the story (argument path, the checkpoint in production/session-state/active.md, or the in-progress stories listed by the template's story-facts.sh script), loads what the story is checked against for its tier (GDD acceptance criteria, ADR decisions, the TR registry, the control manifest), and verifies each acceptance criterion: existence checks for named files and tests (a test file's presence is reported, never its pass — this skill runs no test), questions for behaviour criteria (Yes passes, No fails, Not tested yet keeps the verdict at best NOT ASSESSED), and retained screenshots for visible criteria. It checks the test evidence its type and qa.level require, scans for deviations from the GDD, the ADRs and the story's own scope (against git status --porcelain), and runs or records the skip of the review gates (in full review mode it hands QL-TEST-COVERAGE and LP-CODE-REVIEW to the template's qa-lead and lead-programmer agents, via references/review-gates.md). It then shows the report and asks "How do you want to proceed?" before any close write: on a close choice it sets the story's Status to Complete with Completion Notes, updates production/sprint-status.yaml (silently, covered by that same answer, per its CONTRACT.md), optionally adds one tech-debt row per advisory deviation to docs/tech-debt-register.md when the user picks that option (references/tech-debt-row.md), and prints a suggested git commit (it does not commit). Verdicts: COMPLETE, COMPLETE WITH NOTES, NOT ASSESSED, BLOCKED — closing a blocked or unassessed story is always prompted as a scope decision, in every automation mode. On every run, whatever the verdict and whether or not the story closes, it overwrites the checkpoint block in production/session-state/active.md via the template's session-state.sh script — that write happens without a separate ask and is disclosed here and in the prerequisites. It resolves its many settings by running the template's hooks/yaml-helper.sh, outside this skill's folder, and its CONTRACT.md and three reference files spell out its inputs, outputs and hard constraints. Repository: https://github.com/Donchitos/Claude-Code-Game-Studios (the skill is the whole .claude/skills/story-done/ folder at the pinned commit: 5 files in total) Pinned version: install the code at commit 05ad7c8a5b8aa2aa86bbd3439ae7cf7295cac418 exactly. Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Clone https://github.com/Donchitos/Claude-Code-Game-Studios into a temporary folder, check out commit 05ad7c8a5b8aa2aa86bbd3439ae7cf7295cac418, and copy the .claude/skills/story-done/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-22: [FINGERPRINTS] - CONTRACT.md: SHA-256 b805929ac2ad562172f5ff9d0492268b172cd7db0f3043a94bf35dc7eec9117a - SKILL.md: SHA-256 3946b6770a8f27fd679a0a2f5d9c8aec6096bba6ee7982827700d1b542c9e74f - references/review-gates.md: SHA-256 49618b4e574a5db5dd9b25bc681b92d357b513ef9e05c760a4509d0471913741 - references/sprint-next.md: SHA-256 d18331c32a0e15ac6858d60ac4c9b220b81da607c3f5c83f69eaa691cc932c10 - references/tech-debt-row.md: SHA-256 9bc9090f96ee6b1e3e03a99fee382dcd5029647f07ce1bcdb6d730a492f8c31e If a file does not match, it changed since the listing was prepared: do not install it yet, tell me what you found and let me decide. 2. Read SKILL.md in full, and tell me in plain words: what the skill does when used, what it connects to, whether it reads or sends any personal information or credentials, and anything that seems unusual. Stop on any red flag and tell me before going further. Confirm these points from the files: - The skill folder holds 5 files: SKILL.md, CONTRACT.md (its handoff contract: inputs, outputs, immutability rules and hard constraints) and 3 files under references/ (review-gates, sprint-next, tech-debt-row). Nothing in the folder is a program; the files are instructions and contracts the skill reads and follows at use time. - Nothing is sent anywhere during installation itself; the only download at install time is the skill folder from GitHub at the pinned commit, and no project of mine is audited, read or modified. - Integrated-system reality, from the files: at use time the skill runs the template's hooks/yaml-helper.sh (many resolved keys) and template scripts .claude/scripts/story-facts.sh, story-status.sh, adr-status.sh, md-section.sh, review-receipts.sh and session-state.sh, reads git status, follows .claude/docs/director-gates.md, coding-standards.md and automation-modes.md, reads the story, its GDD, ADRs, TR registry, control manifest and evidence files, and in full review mode spawns the template's qa-lead and lead-programmer gate agents — hooks, scripts, docs and agents that live outside this skill's folder and are absent in a standalone install, so it cannot run as designed on its own. - At use time the story file and sprint-status.yaml are written only on the explicit close choice described above (the yaml rides silently on that same answer), the tech-debt register only on its own option, and a commit is only suggested, never made; the one write that needs no separate ask is the session checkpoint block in production/session-state/active.md, overwritten on every run by session-state.sh and disclosed here. It never writes to the code root or the test root, and no third-party service, account or credential is named in the files. - Do not close, verify or update any story, do not run git status, any template hook or script, and do not use the skill in any way during this install: using it is a separate step that only starts when I ask for it later. 3. When I say "yes, install", install it as a skill: copy the whole story-done folder into the agent's skills directory, in a folder named exactly "story-done", keeping its 5 files in their original relative layout (SKILL.md at its root). 4. Verify with no network calls: frontmatter valid (name: story-done), the 5 files in place with matching fingerprints. Do not run anything from the folder and do not use the skill to verify. 5. Report what was installed, where, and what I still need to do myself (for example: that it belongs to the full Claude Code Game Studios template and may not run standalone (it resolves its settings through the template's yaml-helper hook and works through its .claude/scripts/ helpers and gate agents), and that at use time it closes a story only on my explicit choice, while its session checkpoint in production/session-state/active.md is written on every run without a separate ask, as disclosed in this prompt). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.