5kahoisaac

Analyze, label, and respond to workplace psychological manipulation patterns specific to Varadise and similar tech/startup environments. Use this skill when a user describes experiences at Varadise or similar Hong Kong technology companies involving coercive encouragement, unrealistic workloads, unfair task assignment, extreme micromanagement, blame-shifting, denial of expertise, or pressure to accept exploitation as normal.

A workplace PUA analysis skill for Varadise and similar environments. Use this skill when analyzing coercive management, unrealistic workload pressure, micromanagement, blame-shifting, or other psychological manipulation patterns.

分析、標記並回應職場 PUA（心理操控）模式，特別針對 Varadise 及類似科技／初創環境。當使用者描述在 Varadise 或類似香港科技公司 遭遇被「鼓勵」接受不合理工作量、不公平分工、極端微管理、甩鍋推責、 否定專業意見、或把剝削視為常態時，啟用本技能。

Ingest user-provided files or folders into historian memories by extracting content, classifying it against available memory types, and storing useful knowledge through @historian. Use when the user wants to turn documents, screenshots, notes, code, or mixed source folders into persistent mnemonics.

Memory management by using the historian subagent to store, recall, and manage persistent memories across conversations. Use when you need to remember decisions, preferences, learnings, or retrieve stored context.

Six-phase build discipline that turns one prompt into a complete, runnable deliverable with zero placeholders. Use for ANY non-trivial single-prompt build request — "build me…", "create an app/script/tool/site/game/bot/pipeline that…", "make a complete…", "write a program that…" — and whenever the user says "runnable", "production-ready", "complete project", "end-to-end", "MVP", "one-shot", or invokes the skill by name ("use Fabled", "Fabled workflow", "think like Fable"). Also applies to non-code deliverables (documents, plans, reports, analyses) by mapping the same phases — intent → scope → outline/contract → full draft → verify → deliver. When unsure whether a build request is non-trivial, use this skill.

Use this skill when the user wants to create, update, or plan a skill bundle CSV for the skillless project, especially prompts like "/bundle-skills frontend", "bundle skills for python", "make a react skill pack", "merge these skills into a list", or "find skills and create a category". This skill guides the agent through clarifying the bundle concept, splitting it into useful subgroups, using find-skills/skills.sh metadata, checking install counts and security audit risk, comparing against existing lists/*.csv files, asking whether to merge or create a new bundle, and producing alphabetically sorted CSV rows.

Build polished, sourced shell CLIs (zsh/bash functions) with a consistent house style — TTY-gated colour, glyph fallbacks, y/N confirmation prompts, dependency checks, env-var-with-fallback config, subcommand dispatch, and correct exit codes. Use whenever the user wants to create or improve a command-line tool meant to be sourced from ~/.zshrc or ~/.bashrc (a "gizmo", shell function, dotfiles utility, git helper, port killer, account switcher, updater, etc.), or asks to "add colour", "make it look like a CLI", "add a confirmation prompt", "check for a dependency", or "standardise the output" of an existing shell script. Trigger on requests to write a small CLI as a .sh file, not on application-level programs in Python/Go/Rust.

Add or update a script entry in the "gizmos" catalog README. Use whenever the user adds a new shell script (a .sh file) to their gizmos repo and wants it documented, or says things like "add this script to the readme", "catalog this cli", "update the gizmos readme", "add a section for gac.sh", or drops a script alongside the README and asks to document it. Handles reading the script, generating a standardized section, inserting it in alphabetical order, and rebuilding the table of contents to match. Trigger this even if the user only says "add this to my catalog" with a script file and doesn't mention the readme explicitly.

Extract a PDF/DOCX CV, enhance it with LLM, and transform it into the qwik-resume-editor JSON schema (Resume v2.0.0). Uses markitdown if available, falls back to LLM extraction.

Evaluate experiment results by running the mechanical-evaluator and taste-evaluator agents over every contender in experiments/exp-*/. Use when the user asks to score, grade, evaluate, or rank experiment results, or to (re)generate EVALUATION.md files and a leaderboard. Each contender's result/ is scored to /100 and written to EVALUATION.md in its case folder.