Tech Debt

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. Install "Tech Debt" for me. The skill itself tracks, categorises and prioritises technical debt across a game project's codebase, in four subcommand modes — scan, add, prioritize and report — chosen from its argument; with no subcommand it prints usage and stops with verdict FAIL. Unlike the team orchestrators of the same template it spawns no agent at all (its allowed-tools grant has no Agent tool): it works directly, resolving only automation and workflow first by running the template's hooks/yaml-helper.sh, outside this skill's folder, and following .claude/docs/automation-modes.md for every question it asks. In scan mode it searches the code root — resolved per the template's .claude/docs/code-root-resolution.md (src/, Assets/ or Source/ by engine) — for TODO, FIXME and HACK comments, @deprecated markers, duplicated blocks, files over 500 lines and functions over 50 lines, categorises each finding into Architecture, Code Quality, Test, Documentation, Dependency or Performance debt, and reports the denominator with the findings: if the code root is unresolved it stops as NOT ASSESSED, and if it holds zero source files it reports NOT ASSESSED — no source files to scan, writes nothing to the register and emits no COMPLETE, because zero findings over zero files is no information. Otherwise it presents the findings and asks "May I write these findings to docs/tech-debt-register.md?"; on yes it updates the register without ever overwriting it, matching before appending (a finding with the same file and debt kind updates the existing Open entry, and an Open entry whose pattern has vanished is proposed as Resolved in the same approval), and on no it stops BLOCKED. In add mode it collects a manual entry — description and files in plain text, then category, fix effort (S, M, L or XL) and impact if left unfixed (Low, Med, High or Critical) via AskUserQuestion — applies the same match-before-appending rule, and appends or updates only after asking; the file notes the skill is also preloaded by the template's lead-programmer agent, and in that subagent context, where no AskUserQuestion exists, it returns the three questions in one reply and stops instead. In prioritize mode it reads the register (a missing register is NOT ASSESSED, with a pointer to run scan first), scores each item impact divided by effort, re-sorts, recommends items for the next sprint (or the build order at minimal workflow), presents the result and writes it back only after asking. Report mode is read-only: it reads the register (missing means NOT ASSESSED), summarises open items by category and total effort, items added and resolved in the current sprint, the trend, and flags items sitting in the register for more than three sprints, writing nothing. The register itself follows a fixed ten-column format (TD-NNN ids, never reused) that this skill and the template's /story-done both write. Its frontmatter pins model: sonnet. Repository: https://github.com/Donchitos/Claude-Code-Game-Studios (the skill is the whole .claude/skills/tech-debt/ folder at the pinned commit: 1 file 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/tech-debt/ folder from that checkout. Then compare every file in the folder with the fingerprints recorded for this listing on 2026-10-22: [FINGERPRINTS] - SKILL.md: SHA-256 11299c8a170286b24a040e965fc19d74d9d631b32a021678286e4480a24d1408 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 1 file: SKILL.md only. Nothing in the folder is a program; it is instructions the agent 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 file: at use time the skill runs the template's hooks/yaml-helper.sh to resolve automation and workflow, follows .claude/docs/automation-modes.md and code-root-resolution.md, and reads and greps my project's code root (src/, Assets/ or Source/ by engine) and its docs/tech-debt-register.md locally. It spawns no agent and calls no other skill — but the register it maintains is shared: the template's /story-done writes debt rows into the same file in the same ten-column format, and its Next Steps point to /sprint-plan, so outside the full template the register format and those handoffs have no counterpart. Docs and conventions that live outside this skill's folder are absent in a standalone install, so parts of it cannot run as designed on their own. - At use time the skill writes only the debt register, docs/tech-debt-register.md, and only after asking in each writing mode ("May I write these findings…?", "May I append this entry…?", "May I write the re-prioritized register back…?"); a no stops it BLOCKED, report mode writes nothing at all, and a scan over zero source files writes nothing and reports NOT ASSESSED. The register is updated, never overwritten: duplicates are matched and updated, and vanished patterns are proposed as Resolved inside the same approval. There is no bounded write exception and no silent write in this file. No third-party service, account or credential is named in the file. - Do not scan my codebase, do not add, prioritize or report any debt entry, do not run any template hook, 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 tech-debt folder into the agent's skills directory, in a folder named exactly "tech-debt", keeping its 1 file in their original relative layout (SKILL.md at its root). 4. Verify with no network calls: frontmatter valid (name: tech-debt), the 1 file 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 only automation and workflow through the template's yaml-helper hook, resolves the code root through the template's code-root-resolution doc, and shares its debt register format with the template's /story-done), and that at use time it writes only docs/tech-debt-register.md and only after asking in each writing mode — report mode and empty scans write nothing). 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.