Triage a GitHub issue against the current fallow codebase and determine validity, priority, scope, and likely implementation shape. Use when the user asks to triage an issue, assess a bug report, or evaluate an enhancement request.

Use when pulling changes from rstudio/vscode-server into Positron — fetching upstream, triaging commits, and applying diffs

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

Buy a company or sell yours: diligence, valuation, terms, integration. Use whenever M&A or an exit comes up.

Health check of agent auto-memory: line counts, topic files, capacity, stale entries and what to prune. Use whenever the user asks how full or healthy memory is, or runs /si:memory-status.

Design, deploy and cut the cost of Google Cloud setups: Cloud Run, GKE, Functions, Cloud SQL, BigQuery, moving to GCP. Use whenever the user asks how to host or run something on GCP, why their GCP bill is high, or plans a GCP migration.

Use when a founder question needs routing to the right C-suite advisor, or a cross-functional decision needs a board meeting convened. Loads company context, synthesizes the answers and tracks decisions.

Use when moving tests from Cypress or Selenium to Playwright ('migrate tests', 'convert tests', 'replace selenium').

Terse mode: drop filler, articles and pleasantries, keep technical accuracy, about 75% fewer tokens, and stay in it. Load whenever the user asks for fewer words, shorter answers, less tokens, caveman mode or /caveman.

Use when starting a C-suite advisor session, when company context is missing or older than 90 days, or before company data goes to an external service. Loads and enriches ~/.claude/company-context.md and applies anonymization rules.

Use before analysis or model training on any dataset, or when asked to profile data or hunt missing values and outliers. Profiles it with data_profiler.py, missing_value_analyzer.py and outlier_detector.py and writes a remediation plan.

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

长篇网文规划与写作。支持只讨论结构、只写大纲或指定细纲，明确要求正文后再写章节。触发方式：/story-long-write、/写长篇、「帮我开书」「定设定」「出卷纲」「规划剧情」「写大纲」「补细纲」「日更」「续写」「继续写」「修改第X章」「回炉」「重写第X章」。

Reddit account onboarding and a staged 30-day warm-up workflow through BrowserAct CLI. Use for Reddit warmup, daily browsing and engagement, account handover, reviewing pending comments/posts, and configured brand-promotion workflows. Requires a logged-in Reddit account, a persistent browser profile with a fixed proxy, and an agent that can manage local state and follow approval policies. The skill defines agent workflows; daily scheduling is configured separately.

Search public WeChat articles by keyword through the BrowserAct API, with an article limit and a publication-date value or explicit unrestricted value, returning article metadata, source links and available full text. Use for WeChat article discovery, content research and monitoring.

Fix MSBuild incremental/no-op builds: 'nothing changed but it rebuilds anyway'. USE FOR: targets re-executing on subsequent builds, missing Inputs/Outputs or FileWrites, volatile output paths, untracked writes, glob changes, Visual Studio Fast Up-to-Date Check (FUTDC); 'Building target completely' vs 'Skipping target' in binlog. DO NOT USE FOR: within-build duplicates (find-repeated-work-within-one-build), first-build slowness (build-perf-baseline), parallelism (build-parallelism), slow evaluation (eval-performance), non-MSBuild systems.

Establish MSBuild/.NET baselines before optimizing. USE FOR: unranked slow builds; cold/warm/incremental/no-op measurements; before/after comparisons; CI caching; static graph, artifacts output, dependency graph trimming. DO NOT USE FOR: measured bottlenecks (build-perf-diagnostics), known repeated restores (eliminate-redundant-restores), within-build duplicate calls (find-repeated-work-within-one-build), non-MSBuild builds.

Use when asked to review and fix an open pull request this session did not write — one another tool such as Codex opened, or another session left — and bring it to merge-ready on its own branch. Covers finding and reusing the branch's existing worktree, reading what the branch would run before running it, rebasing it and regenerating bun.lock with evidence instead of merging it, running prep-pr's review lenses on someone else's work, telling a red check main caused from one the branch caused, fixing every finding on the branch, and rewriting the body with write-pr.

Use when starting a piece of feature or fix work in this repository — taking or opening its GitHub issue, cutting the branch (a worktree on the local bare repo, an in-place branch on the remote environment) and producing the implementation plan, including the wiki pages the work will make stale and its observability plan. Also the step for turning a tracker finding into a branch without publishing it.

Use when an issue needs its complexity declared — proposing a rating from the issue body alone and getting the owner to confirm or amend it, or backfilling a rating onto an existing unrated issue. Invoked by write-issue on every issue it opens and by new-feat on an unrated issue it picks up; also the backfill path for issues opened before the label existed.