Skill reviews

Design and build a deck as a python-pptx program: your own layout system, the source's own figures, charts you draw, and a template's design when the user gave one.

• References files that are not bundled: deck/build/references/tables.md, deck/build/references/design-requirements.md, deck/build/references/layouts.md
• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• 15 top-level sections: this looks like several domains in one skill

Verify apps/mobile behavior end-to-end on the iOS simulator against a local API and relay, with Maestro for input and a stand-in host instead of a second machine. Use when asked to

• The skill tells the model to perform an irreversible action with no human approval
• Does not say what the result is
• No condition that starts the skill

Run apps/mobile on the iOS simulator against the local stack, drive it with Maestro, and capture clean screen recordings. Use when asked to record, film, screenshot, or demo the mo

• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (7 tags): a typed call is more reliable
• Does not say what the result is

Operate the user's real desktop apps and windows on macOS, Windows, or Linux. Use when the user asks to open or drive a native app, click or type in a desktop UI, inspect or arrang

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• The skill tells the model to perform an irreversible action with no human approval
• Does not say what the result is

Create a database migration with Drizzle on a fresh Neon branch, and check it is safe to run against production under load. Use when changing the packages/db schema, generating mig

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (3 tags): a typed call is more reliable

Build and publish a self-contained HTML page to Paradev (formerly Superset), then answer the comments readers pin to it. Use this instead of publishing a Claude artifact whenever t

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• 12 top-level sections: this looks like several domains in one skill
• The skill tells the model to perform an irreversible action with no human approval

Draft or update `GUIDELINES.md` at the repo root — the project's full engineering reference. Layer 1 Constitution trinity member (alongside `WORKFLOW.md` universal philosophy and `

• References files that are not bundled: templates/guidelines.md
• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• The skill tells the model to perform an irreversible action with no human approval

Compact the current session into a handoff document a fresh agent can pick up from. Saves to the OS temp dir (never the repo). Captures live state — current branch, open artifacts,

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (8 tags): a typed call is more reliable

Scaffold a staged spike with golden fixtures per WORKFLOW.md §14, for cases where the spec is clear but the technique is uncertain across multiple plausible approaches. Four stages

• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (13 tags): a typed call is more reliable
• Does not say what the result is

Draft or update a Product Requirements Document at doc/product/PRD.md (single-product) or doc/product/<product-slug>.md (multi-product). Layer 3 of the six-layer artifact stack (Co

• References files that are not bundled: templates/prd.md
• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (12 tags): a typed call is more reliable

Use when setting up integration testing for a frontend-backend API boundary, evaluating whether two services are safe to deploy independently, or replacing slow end-to-end tests wi

• The skill tells the model to perform an irreversible action with no human approval
• Does not say what the result is
• No condition that starts the skill

Use when auditing public documentation portals, API references, help centers, SDK docs, or large knowledge bases. Check the final file served at `/llms.txt` and verify that the lin

• The skill tells the model to perform an irreversible action with no human approval
• Does not say what the result is
• No condition that starts the skill

Estimates intracellular metabolic fluxes from steady-state carbon-13 isotope-tracing measurements using validated atom maps, mfapy isotope simulation, constrained multistart fittin

• The skill tells the model to perform an irreversible action with no human approval
• Does not say what the result is
• No condition that starts the skill

Reviews animation and motion code against a high craft bar derived from Emil Kowalski's design engineering philosophy. Default to flagging; approval is earned.

• The skill tells the model to perform an irreversible action with no human approval
• No condition that starts the skill
• Does not say what the process needs to start

How to write modern Swift well — modeling with value types, Swift 6 data-race safety and approachable concurrency (@concurrent, main-actor-by-default, actors, task groups), protoco

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• 18 top-level sections: this looks like several domains in one skill
• The skill tells the model to perform an irreversible action with no human approval

The response, error, pagination, and validation contract every `/api/v2` endpoint must satisfy. Use when adding or changing a route under `apps/sim/app/api/v2/`, or when auditing o

• References files that are not bundled: ../../../apps/sim/lib/api/contracts/v2/openapi/README.md
• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• 12 top-level sections: this looks like several domains in one skill

Add a runtime feature flag (AppConfig-backed on prod, secret fallback off-prod), global by default or optionally gated by workspace id, org id, user id, or platform admin

• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (5 tags): a typed call is more reliable
• Does not say what the result is

Validate an existing Sim webhook trigger against provider API docs and repository conventions

• 10 top-level sections: this looks like several domains in one skill
• The skill tells the model to perform an irreversible action with no human approval
• Does not say what the process needs to start

Author or review a Drizzle DB migration for zero-downtime safety — expand/contract phasing, backward-compatibility with the deployed app version, and writing the `-- migration-safe

• References files that are not bundled: scripts/check-migrations-safety.ts
• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• The skill tells the model to perform an irreversible action with no human approval

Commit, push, and open a PR to staging in one shot — runs the cleanup pass and, when migrations changed, the db-migrate safety review first

• References files that are not bundled: row|chunk|document|vector|tuple|doc, CONTRIBUTING.md, scripts/*.test.ts
• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• The skill tells the model to perform an irreversible action with no human approval

test on Windows, enterprise CA, corporate certificate, GPO cert, TLS fetch failed, Windows sandbox, daytona windows, self-hosted cert. Use when validating OpenWork Windows enterpri

• References files that are not bundled: scripts/support/setup-openwork-tls-repro.ps1, scripts/support/openwork-doctor.ps1
• The skill tells the model to perform an irreversible action with no human approval
• 2 referenced file(s) missing: scripts/support/setup-openwork-tls-repro.ps1, scripts/support/openwork-doctor.ps1

Boot, reopen, update, or reset OpenWork PR previews. Discover script worlds; choose preview-desktop (app only), preview-den (Den only), preview-full (Den plus desktop), or preview-

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (13 tags): a typed call is more reliable

Deploy a Dockerfile, source folder, or existing image to Cloudish as a running container at a live URL, built server-side with no local Docker, with confirmation before spending cr

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• The skill tells the model to perform an irreversible action with no human approval
• Does not say what the result is

Triage the improvements backlog at .turbo/improvements.md: merge duplicate entries, state the user story or simplification behind each one, and decide with the user which to keep a

• The skill tells the model to perform an irreversible action with no human approval
• Does not say what the result is
• Does not say what the process needs to start

Open-source pipeline: fork, sanitize, and package private projects for safe public release. Chains 3 agents (forker, sanitizer, packager). Triggers: '/opensource', 'open source thi

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• The skill tells the model to perform an irreversible action with no human approval
• Does not say what the result is