Comentaris
Comentaris i anotacions del codi
argent-lens
thorgas
Propose multiple visual design variants for on-screen elements and let the human pick in the Argent Lens window. Use when the user asks for design alternatives / options / A-B choices for a screen or component, or any time you have produced more than one candidate look for an element and want a human decision before committing.
spec-driven-develop
lidicn
Automates pre-development workflow for large-scale complex tasks. Use when the user mentions "rewrite", "migrate", "overhaul", "refactor entire project", "transform", "rebuild in [language]", "spec-driven", or describes any large-scale project transformation that requires planning before coding. Also triggers on Chinese keywords: "改造", "重写", "迁移", "重构", "大规模", "规范驱动". Performs full project analysis, task decomposition, documentation generation, project-level instruction and native memory surface resolution, progress tracking setup, and then executes the plan within the same session.
simplify
thorgas
Keep code, tests, documentation, configuration, and other changes as tight, clear, performant, and minimal as possible without weakening behavior. Use during implementation to prevent unnecessary complexity, after behavior is proven to consolidate working changes, before handoff for a full branch audit, and when asked to simplify, tighten, clean up, reduce verbosity, remove complexity, or minimize existing work.
review-axiom-task
wheakerd
Review an Axiom task's routing, scope, authorization, actions, evidence, stops and outcome; explain, audit, criticize, appeal or narrow its decisions, including why it selected, allowed or refused an action. Ordinary summaries, code review, execution and retry stay outside. Clarify mutually exclusive review versus execution/retry before routing; review grants no action authority.
50-coverage
gamesexplained
The annotation loop. Measure comprehension with the shared coverage metric, work the burn-down queue of largest undescribed runs, export symbols.json after every session. Includes how to split the work across subagents safely.
tool-regen2000
gamesexplained
How to drive regenerator2000 on a BBC Micro game. Starting it on the kit's jsbeeb snapshot in its BBC Micro mode, how the BBC's text, the MOS's names and SHEILA's registers come through to symbols.json and the listing, and the traps that are the BBC's. The rest is the C64's skill.
tool-regen2000
gamesexplained
How to drive regenerator2000, the recommended disassembler, through its MCP server with the kit's client script. Starting it on a snapshot, the tools that matter, the annotation log, and the traps.
40-sweep
gamesexplained
Cheap mechanical sweeps that convert large regions from unknown to understood in one pass (a hardware register census, a string sweep, a twin-copy check and, when another version of the game is documented, a map of that version onto this image). Also how to start from an analysis the contributor already has. Run early, and again whenever the burn-down stalls.
70-minisite
gamesexplained
Build the game's minisite. The "How it works" page (index.html) from symbols.json, facts.md and the reference images; the Discoveries tab of secrets, quirks and bugs; the listing behind the Source code tab; the Maps / levels tab for a game with more than one level; an optional Play tab. Interactivity first and the best of it high on the page, evidence beside every claim, copy written last under the house style.
60-verify
gamesexplained
Turn traced claims into verified facts. Test cheaply testable claims live in the emulator, hunt down the negative results you are relying on, and write facts.md as current truth with evidence.
tool-skoolkit
gamesexplained
How to drive SkoolKit 10.1 for ZX Spectrum disassembly — the control file that is the annotation surface, the directives the kit writes, the round trip through symbols.json, and the traps. The kit's own Z80 decoder and snapshot reader mean a listing never depends on it. The machine facts it needs are in zx-spectrum-reference.
review-round
AME-Team
Dual-Gate レビューラウンドを自動化する。Gate 1（pre-commit）と Gate
review-game-pr
gamesexplained
Review a gamesexplained pull request that adds or updates a game against the maintainer's checklist, and recommend the next step. Run by hand on a PR number, or by the hourly routine below.
mdpreview
tdimino
"Catppuccin-themed live-reloading Markdown viewer with multi-tab support, margin annotations, sidebar TOC, and Chrome --app mode. This skill should be used when previewing .md files in a polished editorial reader, adding AI or human annotations to documents, or launching a collaborative document review session."
typescript-satisfies-operator
flpbalada
Guides proper usage of TypeScript's satisfies operator vs type annotations. Use this skill when deciding between type annotations (colon) and satisfies, validating object shapes while preserving literal types, or troubleshooting type inference issues.
comment-mode
shipshitdev
Granular feedback on drafts without rewriting. Generates highlighted HTML with click-to-reveal inline comments. Use when user says "comment on this", "leave comments on", "give feedback on", or asks for feedback on a draft. Supports multiple lenses—editor feedback, POV simulation ("as brian would react"), or focused angles ("word choice only", "weak arguments"). A granular alternative to rewrites that lets users review feedback incrementally without losing their voice.
consolidate-task-prs
osdlc
Collapse every open single-task-file PR — each carrying one new draft task file under docs/planning/tasks/ — into a single consolidation PR, deduping the folded tasks against the present task list, then close the superseded PRs and delete their branches. Explicit cleanup only, never auto-invoked: run it to fold an accumulation of one-task-each PRs back into one reviewable PR.
code-review
osdlc
"Review a change for correctness bugs and report concrete findings with a fix for each. The rubric other reviewing skills embed: what to read, what to look for, what to skip, and how to write a finding."
comment-review
osdlc
"Review the comments in a diff or a subtree and minimize them against the expression ladder: a fact belongs first in the identifier, then in the signature and types, then in the declaration's doc annotation, and only when none of those can carry it does it earn an inline comment. Deletes comments that restate the code, moves facts down the ladder, tightens what survives to one short literal statement, strips volatile project state, and replaces duplicated prose with a reference to one canonical home. Invoke for \"review the comments\", \"too many comments\", \"comment cleanup\", or as the comment pass on a branch before review."
review-pr
osdlc
Review one or more pull requests and post the findings as a GitHub review. A marker in the posted review records the head commit it covered, so a later run with nothing new either stops (--mode auto) or asks whether to review again (--mode ask, the default), and a run over new commits reviews the delta since the last review first while still judging the changeset as a whole. Keeps its state on a per-PR git ref on origin (claim, findings, answers), rebuilt from the PR when missing, so several sessions can run the same sweep without posting twice. Suits a loop such as /loop /review-pr --mode auto.
parley
osdlc
Check, serve and render a parley document as one review page, then read the reviewer's comments and answers back and reply to each one. A document is JSON: a list of typed blocks and the questions the agent needs answered. The peer parley.mjs is one bundled file that runs under bun, node 22+ and deno with no dependencies and no build step.
deslop
strikersam
Remove AI code slop before committing. Review AI-generated code for unnecessary verbosity, redundant comments, over-engineering, and low-signal patterns that degrade codebase quality over time.
cron-scheduler
AIDotNet
解析、验证和解释cron表达式，计算下次运行时间。
documentation-writer
AIDotNet
生成全面的文档，包括README文件、API文档、代码注释（JSDoc、docstrings、XML）和架构文档。