Skill reviews

Full PR lifecycle in a fresh task-owned git worktree: implement via the ulw-loop skill with mandatory evidence-bound manual QA → reviewer-readable English PR → verification loop (C

• 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 (9 tags): a typed call is more reliable

Write a handoff file at .turbo/handoff/<YYYY-MM-DD>-<slug>.md capturing current session state — task, status, open decisions, in-flight changes, next step — so a fresh session can

• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (6 tags): a typed call is more reliable
• Does not say what the result is

Prepare and publish DeepChat releases in this repository. Use when Codex needs to bump the app version, update CHANGELOG.md, keep release notes bilingual from v1.0.1 onward with En

• References files that are not bundled: ../../../docs/release-flow.md
• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (9 tags): a typed call is more reliable

Curated upstream guidance for Apify Integration Development; use when the workflow matches the user goal.

• 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

Register a finished skill in this library and produce the complete release package: the bundle wiring and marketplace entry, the regenerated catalogue and exports, the headline cou

• References files that are not bundled: scripts/new-bundle.mjs
• 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

Use when rendering a Markdown document (English or Korean) such as a proposal, IRB cover letter, handout or reference table to PDF via pandoc and xelatex, with auto-fitted table wi

• References files that are not bundled: scripts/render_pandoc.sh, scripts/xecjk_korean.tex
• The skill tells the model to perform an irreversible action with no human approval
• 2 referenced file(s) missing: scripts/render_pandoc.sh, scripts/xecjk_korean.tex

Use when running a systematic review and meta-analysis, DTA or intervention. Covers PROSPERO protocol, search, screening, extraction, risk of bias (QUADAS-3, RoB 2, ROBINS-I), biva

• 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 (4 tags): a typed call is more reliable

Use when preparing an academic talk such as a journal club, grand rounds, seminar, conference presentation, or lecture/teaching deck. Analyzes the source, drafts audience-adapted s

• 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 (6 tags): a typed call is more reliable

Use when a manuscript or response-to-reviewers letter reads as AI-written. Scans for 27 AI writing patterns and rewrites flagged passages, preserving technical accuracy and boundin

• References files that are not bundled: scripts/check_rhetorical_density.py, references/r2r_voice.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

Publication-grade correctness and legibility rules for final-deliverable scientific figures, not exploratory plots. Use for a figure that will ship in a report, paper, export, or k

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• 11 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

Manage ownership, retention, and cleanup of downloaded weather data across one-shot jobs, interactive viewers, caches, failures, and cancellation.

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• 11 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

Rules for when an agent should recall from long-term memory before acting and when it should save decisions, corrections and failures afterwards. Works with any memory backend.

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• 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

Prove that a command-line tool actually did what it said, instead of trusting its exit code and its output. Snapshots the filesystem before and after, names the expected consequenc

• 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
• Does not say what the result is

Discover and access NEXRAD data for a selected radar site, time, product, or Level II moment using completed archive volumes, real-time chunks, or supported Level III sources with

• 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
• The skill tells the model to perform an irreversible action with no human approval

内容创作者与运营专员让 AI Agent 建立专属社交人设时，当用户明确要求 Agent 进行社交互动或社区运营时使用。一键完成注册、发帖、评论、点赞及创建社区，轻松打造 AI 专属社交圈。

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• 25 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

Design and build waiting-list portals for anticipated goods or services: classify whether the experience needs interest capture, verified early access, referral growth, a virtual w

• The skill tells the model to perform an irreversible action with no human approval
• No test case covers injection arriving through data
• Does not say what the process needs to start

Land a change everywhere the same fact is stated — enumerating the full surface inventory (landing copy, docs, machine-readable summaries, changelog badges repeated across every pa

• 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

Establish a finding before you publish it, and correct it after — headlines that overstate what actually reproduces at the layer a user sees, reporting code that no entry point can

• 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 result is

Write or tighten the description field of a skill or agent so the right one loads. Use when creating a skill or agent, when a description is vague, bloated, or missing triggers, wh

• The skill tells the model to perform an irreversible action with no human approval
• 8 test cases, all positive: not one "should refuse" or "should ask first"
• No test case covers injection arriving through data

Reference guide for the Twelve-Factor App methodology — 15 principles (12 original + 3 modern extensions) for building portable, resilient, cloud-native applications. Use when eval

• 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

Start or complete a specific refactoring task from a task file. Use when a sub-agent needs to pick up a refactoring task, update its status, implement acceptance criteria, and run

• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (4 tags): a typed call is more reliable
• No condition that starts the skill

Optimize existing CLAUDE.md, SKILL.md, agent definitions, and other AI-facing files for Claude comprehension and economy. Scope: optimization of existing content only — not upstrea

• 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

Use this skill when building, enriching, auditing, or uploading B2B paid ads audiences to Google Customer Match, Meta/Facebook Custom Audiences, or LinkedIn Matched Audiences. Trig

• Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
• 17 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

Build, publish, and verify a Deepline Play as a durable state machine over Customer DB tables, including a small paid pilot. Invoke this skill explicitly when the user asks to buil

• The skill tells the model to perform an irreversible action with no human approval
• The response is described with custom markup (9 tags): a typed call is more reliable
• Does not say what the process needs to start

Track what top crypto futures traders hold on Binance, Hyperliquid, Bybit and OKX with TraderSpy: elite leaderboard, positions, trader profiles and history. Use for "what are whale

• 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