Comments
Code comments and annotations
vdjdb-extract
by antigenomics
Extract TCR:pMHC specificity data from raw source files (papers, supplementary tables, XLS, PDF, 10X output, AIRR-format) and produce a VDJdb-formatted TSV chunk ready for /format and /proofread.
staged-spec-review
by phazei
Review one stage of a multi-stage specification against the full plan to catch forward-compatibility gaps before they compound. Use whenever the user asks for a review of a single phase, prompt, RFC section, migration step, or implementation that is part of a larger sequence — especially when they provide both the artifact under review AND the full multi-stage spec. Triggers on phrases like "review prompt N against the full spec", "check this phase against the others", "did stage 1 set up correctly for stages 2-5", "is this RFC consistent with the rest", or any request to audit a single piece of a sequenced plan with the rest of the plan as context. Use this even when the user doesn't explicitly say "staged" — if the structure is clearly sequential (prompt 1 of 5, phase A of D, migration step 2, etc.) and they're asking about one piece, this skill applies.
Douban
by imoonkey
```
workflow-pr-doctor
by cuioss
Diagnose and fix PR issues (build, reviews, Sonar) — thin wrapper around verification-feedback with producer=pr-state
paperproof-protocol
by PaperProofLabs
Operate the PaperProof protocol for AI agents and developer assistants. Use when a user asks to publish, update, query, verify, inspect, or explain PaperProof artifacts, versions, comments, governance, native prompts, Agent Memory registrations, Walrus content, Sui package/object bindings, wallet readiness, or SDK-based PaperProof workflows. Prefer protocol, SDK, Sui, Walrus, and indexer interfaces over browser automation or website clicking.
write-discoverable-code
by SjB
Rules for writing code that coding agents (and humans) can find and understand through plain-text search. Apply whenever writing or renaming code: functions, types, constants, files, error messages, doc comments. Grounded in measurement: agents navigate by plain-text search, not by AST or language server, so every identifier is a search query and every search miss costs wasted reads.
respond-to-eval
by pedrohcgs
Turn student course evaluations (free-text + numeric) into an actionable teaching-improvement plan — the teaching analogue of /respond-to-referees. Clusters comments into themes, separates signal from noise, classifies each theme Keep / Change / Investigate / Out-of-scope, and drafts concrete changes mapped to the syllabus and slide decks. Use when user says "respond to my evals", "what do these course evaluations tell me", "turn my teaching feedback into a plan", or after a semester's evals arrive.
code-upkeep
by Sylla-BV
This skill should be used when the user asks to "update docstrings", "audit docstrings", "fix docstrings", "add docstrings", "check docstrings", "Google-style docstrings", "add field descriptions", "update tests", "audit test coverage", "generate tests", "add tests", "fix tests", "write pytest tests", "check test coverage", "missing tests", "stale tests", "test this module", "add test cases", "bring code up to date", "code upkeep", "maintain code", or wants to audit and update Python docstrings or tests.
research-refactor
by qijialiv-cell
"Research code refactoring framework for R/Stata/Python projects. Enforces safe, incremental, auditable refactoring with dependency tracking and atomic commits."
issues-workflow
by terrylica
Plan and track work using a GitHub Issues-first workflow with sub-issue hierarchies, issue-branch-PR lifecycle, and auto-close on merge.
ocm-sdk
by andreibesleaga
How to use the ocm SDK. For access to more information, try out the ocm MCP server.
docx-tracked-changes
by mohitagw15856
"Produce properly-formatted tracked changes for a Word document. Use when asked to redline a document, suggest edits to a contract or document, create tracked changes for review, or mark up a document with proposed revisions. Produces a complete redline with insertions, deletions, and margin comments that can be applied to the source document. Best used with Claude Opus 4.7 or newer for reliable tracked changes handling."
project-management
by dotnet
'GitHub issue and project-board management for the dotnet/msbuild repo. Use when asked to file/triage/update issues, post comments, amend issue bodies, move sprints, bulk-update project board fields, or audit items by sprint/status/assignee.'
eli5
by scottwater
Explain a topic like I'm a 5 year old with a simple HTML picture artifact.
cursor-agents-workflow
by langfuse
Human handoff, Linear branch names, reviewable (non-draft) PRs, Claude, Greptile, and Codex review comments, preview test steps, proof of work posted on the GitHub PR, and review-doubt notes for Cursor agents. Use when a Cursor Cloud or Cursor desktop agent implements a Linear issue, opens a GitHub PR, asks a human to test, posts screenshots or videos, or handles Claude, Greptile, or Codex code-review comments.
morning-triage
by luxtelos
Discovery — scan GitHub issues (and anything else the project names) for work, rank it, and append rows to state/triage.md at status new. Never modifies or closes existing rows.
yoyo-string-annotation
by EvannZhongg
Build high-quality yoyo string annotation collections by having a vision-capable agent directly inspect, annotate, repeatedly refine, self-review, audit, and export labels without detector, segmentation, pose, SAM, or other recognition-model prelabels. Use for dispersed video frame sampling, thin-string centerlines or masks, normal-versus-horizontal trick orientation, ordered hand-to-yoyo path continuity, adjacent-frame refinement, hard negatives, unresolved-case handling, and source-group-preserving annotation production before any later dataset split.
hung-yi-lee
by n840229
"Explain machine learning, deep learning, generative AI, LLMs, AI agents, and speech modeling in a Hung-Yi Lee-inspired teaching style. Use this skill when the user wants 李宏毅式教學: roadmap-first structure, intuition before math, black-box-to-mechanism explanations, everyday analogies, anticipating student confusion, practical debugging, and research-grounded context."
teachback
by terrylica
Build or deepen a single-file interactive HTML explainer that teaches back difficult material already established in the conversation (or a named subject/source). Use when someone wants a durable, self-contained lesson with rendered mathematics, an exploratory multiple-choice assessment, and a terminology loop that lets the reader push unfamiliar terms back for deeper coverage.
SKILL: harn-bitbucket-connector
by burin-labs
For SVN polling, add a poll trigger and run harn connector check . --provider svn --run-poll-tick.
perfect-landing-page
by da1z
Build high-converting landing pages with a simple 6-section framework. Use for landing page guidance.
ashare-screener
by lc132
A股每日盘前短线标的智能筛选(v6.9.28)。基于前一日收盘数据，通过 38步筛选流程（网络授时北京→GitHub持仓跟踪→节假日→极端行情→外围市场→持仓同步→做T→持仓跟踪→持仓危机→全市场API拉取→行业查表补全→pytdx历史K线→东方财富财务(已降级)→F10单股API→风险事件→拥挤度(机构持仓+融资过热)→13项硬排除→27项信号过滤→十七策略评分→行业集中度→生成HTML报告→GitHub同步→飞书推送→每周复盘），输出短线标的_YYYYMMDD.md 为Markdown格式，同时生成可视化HTML报告。当用户需要运行盘前筛选、A股短线选股、每日标的预测时使用。
Backend Engineering Skills for Doxynix
by doxynix
This philosophy guides all design decisions.
masm-padding
by tomevault-io
Enforce stack padding conventions for Miden Assembly (.masm) procedures based on invocation type (call vs exec). Use when editing, reviewing, or creating .masm procedures, especially those with Invocation annotations. Use when this capability is needed.