odysseythink
@odysseythink
Public Skills
finishing-a-development-branch
by odysseythink
ALWAYS invoke this skill when implementation is complete, all tests pass, and you need to decide how to integrate the work — merge, open a PR, or clean up the branch. Presents structured options and executes the user's choice, e.g. when the user asks "what now", "merge or PR", or "close out this branch".
using-git-worktrees
by odysseythink
ALWAYS invoke this skill when starting feature work that needs isolation from the current workspace, or before executing implementation plans — ensures an isolated workspace exists via native tools or git worktree fallback. E.g. "start this on its own workspace" or "prepare before running the plan".
executing-plans
by odysseythink
ALWAYS invoke this skill when you have a written implementation plan to execute in a separate session with review checkpoints — e.g. "execute this plan", "run the plan in a new session", or a plan file is handed off for another session to carry out. Not for same-session execution; that is subagent-driven-development.
systematic-debugging
by odysseythink
ALWAYS invoke this skill when encountering any bug, test failure, or unexpected behavior, before proposing fixes — e.g. "it's broken", "tests are failing", "something is off". Finds the root cause first; no fix until the cause is proven.
verification-before-completion
by odysseythink
ALWAYS invoke this skill when about to claim work is complete, fixed, or passing — before committing, opening PRs, or telling the user anything is done — and when the user asks "are we done", "is it fixed", or "ready to merge". Runs the verification commands and confirms their output first; evidence before assertions always.
verification-before-completion
by odysseythink
ALWAYS invoke this skill when about to claim work is complete, fixed, or passing — before committing, opening PRs, or telling the user anything is done — and when the user asks "are we done", "is it fixed", or "ready to merge". Runs the verification commands and confirms their output first; evidence before assertions always.
test-driven-development
by odysseythink
ALWAYS invoke this skill when implementing any feature or bugfix, before writing implementation code — whenever new behavior or a fix is about to be written. Writes the failing test first, then the implementation that makes it pass.
using-git-worktrees
by odysseythink
ALWAYS invoke this skill when starting feature work that needs isolation from the current workspace, or before executing implementation plans — ensures an isolated workspace exists via native tools or git worktree fallback. E.g. "start this on its own workspace" or "prepare before running the plan".
systematic-debugging
by odysseythink
ALWAYS invoke this skill when encountering any bug, test failure, or unexpected behavior, before proposing fixes — e.g. "it's broken", "tests are failing", "something is off". Finds the root cause first; no fix until the cause is proven.
test-driven-development
by odysseythink
ALWAYS invoke this skill when implementing any feature or bugfix, before writing implementation code — whenever new behavior or a fix is about to be written. Writes the failing test first, then the implementation that makes it pass.
dispatching-parallel-agents
by odysseythink
ALWAYS invoke this skill when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies — e.g. the user asks to "run these in parallel", "dispatch agents", or "can these be done at the same time". Screens the tasks for true independence and dispatches each with a self-contained brief.
skill-creator
by odysseythink
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Ody's capabilities with specialized knowledge, workflows, or tool integrations.
finishing-a-development-branch
by odysseythink
ALWAYS invoke this skill when implementation is complete, all tests pass, and you need to decide how to integrate the work — merge, open a PR, or clean up the branch. Presents structured options and executes the user's choice, e.g. when the user asks "what now", "merge or PR", or "close out this branch".
executing-plans
by odysseythink
ALWAYS invoke this skill when you have a written implementation plan to execute in a separate session with review checkpoints — e.g. "execute this plan", "run the plan in a new session", or a plan file is handed off for another session to carry out. Not for same-session execution; that is subagent-driven-development.
frontend-design
by odysseythink
Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.
vibedrop
by odysseythink
Deploy any static site or build output to a public URL via VibeDrop. Use when the user asks to share, deploy, publish, host, or preview a web page, landing page, static site, or HTML/CSS/JS bundle.
legal-contract
by odysseythink
Draft and review Chinese legal contracts with an optional built-in critic gate. Produces contract drafts and runs a quality review before delivery.
data-analysis
by odysseythink
Data analysis specialist. Analyzes datasets, identifies patterns, generates insights, and creates summary reports. Use when working with data, statistics, or analytical tasks.
idea-evaluator
by odysseythink
Evaluate the quality of a startup or product idea BEFORE committing to the six-model review. Use this skill whenever the user describes a new startup idea, product concept, or business direction they're considering — especially during the P1 open exploration phase. Also trigger when the user says "I have an idea", "what do you think of this direction", "should I explore X", "I noticed a problem with Y", or presents any concept that could become a product or business. This skill runs 10 evaluation questions, 4 mistake checks, 3 counter-intuitive signal checks, and outputs a structured quality score. The output feeds directly into the six-model review: ideas scoring ≥6/10 proceed to full review, ideas scoring 3-5 get specific improvement suggestions, ideas scoring ≤2 get a recommendation to drop or radically rethink. Always use this skill BEFORE the user starts building, designing architecture, or running the six-model review on a new idea. This is the first gate in the funnel.
idea-evaluator
by odysseythink
Evaluate the quality of a startup or product idea BEFORE committing to the six-model review. Use this skill whenever the user describes a new startup idea, product concept, or business direction they're considering — especially during the P1 open exploration phase. Also trigger when the user says "I have an idea", "what do you think of this direction", "should I explore X", "I noticed a problem with Y", or presents any concept that could become a product or business. This skill runs 10 evaluation questions, 4 mistake checks, 3 counter-intuitive signal checks, and outputs a structured quality score. The output feeds directly into the six-model review: ideas scoring ≥6/10 proceed to full review, ideas scoring 3-5 get specific improvement suggestions, ideas scoring ≤2 get a recommendation to drop or radically rethink. Always use this skill BEFORE the user starts building, designing architecture, or running the six-model review on a new idea. This is the first gate in the funnel.
roadmap-architect
by odysseythink
Turn a large multi-phase roadmap into an execution-ready plan by splitting oversized goals (so an AI executor never silently drops scope), ordering real source-level dependencies, judging what can run in parallel, and labelling each sub-task with the execution mode it should run in (normal / plan / design). Use this skill whenever the user wants to "拆分 roadmap", "细化路线图", "标注每个阶段用什么模式跑", "排依赖 / 看能不能并行", "防止 AI 执行时遗漏功能", or asks which sub-tasks are safe in normal mode vs need plan mode vs need design mode. The defining feature is ROI discipline: it does the high-leverage work once (one shared rubric, applied in a single batch pass over every phase) instead of re-deriving the same judgement phase-by-phase across many turns. Do NOT use this skill to actually execute a roadmap task — it only produces and refines the plan.
roadmap-architect
by odysseythink
Turn a large multi-phase roadmap into an execution-ready plan by splitting oversized goals (so an AI executor never silently drops scope), ordering real source-level dependencies, judging what can run in parallel, and labelling each sub-task with the execution mode it should run in (normal / plan / design). Use this skill whenever the user wants to "拆分 roadmap", "细化路线图", "标注每个阶段用什么模式跑", "排依赖 / 看能不能并行", "防止 AI 执行时遗漏功能", or asks which sub-tasks are safe in normal mode vs need plan mode vs need design mode. The defining feature is ROI discipline: it does the high-leverage work once (one shared rubric, applied in a single batch pass over every phase) instead of re-deriving the same judgement phase-by-phase across many turns. Do NOT use this skill to actually execute a roadmap task — it only produces and refines the plan.
skill-installer
by odysseythink
Install Ody skills into $ODY_HOME/skills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo (including private repos).
debt-ledger
by odysseythink
Scan the codebase for // ody: / # ody: simplification-debt markers and render a Chinese-first ledger report grouped by file with rot-risk warnings. Use after every simplification session to harvest markers, and before code review to surface pending simplifications.