thaitype

@thaitype source repo

31 skills
0 followers
6 hours ago last updated

▌
Typmem Judge · thaitype
Review captured learnings (acting as an LLM-as-judge) in ~/.typmem/memory/learnings/ and decide automatically what to do with each — revise it in place when the fix is cheap and safe, or raise a proposal in ~/.typmem/memory/proposals/ when it should be promoted, conflicts with a higher layer, or is unclear. Use this when asked to "judge", "review the learnings", "vet the inbox", "triage learnings", "clean up captured facts", or after a batch of typmem-learn captures has piled up. This never blocks: learnings are valid and usable whether or not a human ever acts on a proposal. Do NOT use it to write precedents or rules directly (those come from a human acting on a proposal), and do NOT use it on retros (retros are diaries and are never reviewed).

0
installs
▌
Typmem Learn · thaitype
Capture a durable fact — a "learning" / recipe card — to ~/.typmem/memory/learnings/. A learning is a distilled, timeless fact about the world (a gotcha, a constraint, a how-things-work truth) that stays useful long after the work that produced it is done. Use this whenever you discover something worth keeping, when the user says "remember this", "note that", "save this", "TIL", or "lesson learned", and use it to extract durable facts out of a retro after a session. This is the primary memory-capture skill — reach for it often, even when the user doesn't say the word "learning". Do NOT use it for session narrative, status, or what-happened-today; that is a retro — use typmem-retro for those.

0
installs
▌
Typmem Retro · thaitype
Write a retrospective diary entry capturing the story of a work session — what you set out to do, what happened, decisions made, dead ends hit, and unfinished threads — to ~/.typmem/memory/retro/YYYY-MM-DD-<slug>.md. Use this whenever wrapping up work, when the user says they are done / stopping for now / "end of session", when a meaningful chunk of work concludes, or when asked to "write a retro", "log what happened", "record today", or "capture this session". Reach for it even if the user never says the word "retro" — any end-of-work wrap-up is a trigger. Do NOT use this to capture durable facts, gotchas, or how-things-work lessons; those are learnings — use typmem-learn for those.

0
installs
▌
Chief Init · thaitype
Bootstrap `.chief/project.md` for a Chief-installed project by interviewing the user about their tech stack, dev commands, architecture, and key rules. Use after `/chief-install` (or any time the user wants to set up project-wide context). This is the lazy entry point for `.chief/` — it creates only `project.md`; milestones and rules are created later, on demand.

0
installs
▌
Chief Loop · thaitype
Run chief-agent across as many batches as it takes to finish a milestone, writing one report per task instead of one per batch. When a task hits ambiguity, a throwaway decision-support agent proposes options; chief-agent still makes the final call and the report captures the reasoning. Requires goals and contracts to exist. Use "/chief-loop".

0
installs
▌
Chief Plan · thaitype
Plan a new milestone or extend an existing one step-by-step with review gates. Starts with a grill-design session to clarify requirements, then walks through goal → contract → todo → task specs, pausing for user approval at each step.

0
installs
▌
Chief Rule · thaitype
Proactively add a rule to `.chief/_rules/`. Counterpart to `/chief-retro` (which proposes rules reactively from observed patterns). Use when the user has a rule in mind and wants to capture it now. Interviews the user (what / why / when / examples), classifies the rule into a category (_standard / _contract / _goal / _verification), shows a draft, and writes after confirmation. One rule per invocation.

0
installs
▌
Dump Commit · thaitype
Quick commit all files with a short 1-line message. Saves tokens by skipping detailed diff analysis. Use when the user wants a fast save (e.g. "/dump-commit" or "/dump-commit fix upgrade flow").

0
installs
▌
Chief Grill · thaitype
Deep grill that interviews the user one question at a time AND verifies each answer against the codebase via background `answer-verifier-agent` calls. Catches factual conflicts, contradictions with prior decisions, and unfounded assumptions inline. Persists the session to `.chief/_grill/opened/NNNN-topic.md` so it survives context compaction. Use when the user wants a stress-tested grill — when stakes are high, when claims must be cross-checked against actual repo state, or when the topic spans many decisions. Heavier than `/grill-design`; prefer this when correctness matters more than speed.

0
installs
▌
Chief Retro · thaitype
Run a retrospective on the current milestone or latest batch. Checks goal/contract coverage, summarizes planned vs delivered, and proposes rule updates. Use "/chief-retro" after completing a batch or milestone.

0
installs
▌
Chief Install · thaitype
Install the Chief framework into the current project. Installs subagents and AGENTS.md only — `.chief/` is created lazily by chief-agent at runtime. Use when the user wants to set up the framework (e.g. "/chief-install" or "/chief-install canary").

0
installs
▌
Chief Upgrade · thaitype
Upgrade the Chief to a specific version. Uses upgrade.sh as the primary method, falls back to manual if script fails. Use when the user wants to upgrade the framework (e.g. "/chief-upgrade" or "/chief-upgrade canary").

0
installs
▌
Shape Up · thaitype
Co-write a top-down design spec for software projects. Shape vague ideas into clear specs by working layer-by-layer — vision, scope, building blocks, then detail. Use when user wants to design a system, plan a project, write a design spec, or needs to see the big picture before diving into details. Use before /grill-design to prevent losing focus on large projects.

0
installs
▌
Chief Autopilot · thaitype
Run chief-agent in full autopilot. Requires goals and contracts to exist. Chief creates TODO, delegates to builder, and repeats until milestone is done. Auto mode makes all decisions autonomously; safe mode stops on ambiguity. Use "/chief-autopilot" for auto or "/chief-autopilot safe" for safe mode.

0
installs
▌
Slim Down · thaitype
Cut an over-engineered plan down to one achievable increment. Reads the current codebase to understand what exists, then trims scope to the smallest useful step that still scales. Use when user says "slim down", "too much", "over-engineered", "simplify the plan", "what's the MVP", "cut scope".

0
installs
▌
Grill Design · thaitype
Stress-test a design, idea, or approach by interviewing the user one question at a time. Each question comes with a recommended answer AND a one-line self-critique of that recommendation. After each user answer, briefly stress-test the answer against prior decisions before moving on. Use when the user wants a sharp grill on a design or decision tree without persistence overhead. Lighter than `/chief-grill` (which adds codebase verification and a session log).

0
installs
▌
Loop Readiness · thaitype
Review the readiness of a plan for an unattended/autonomous loop — checks whether it has enough feedforward guidance and feedback verification to run safely. Use when the user has a plan (or an idea) for a loop/agent that will run multiple iterations without a human checking every step, and wants to know how well-instrumented it is before starting. Works from an existing plan file or from a live conversation.

0
installs
▌
Time · thaitype
Get the current local time and date, optionally in a specified timezone/UTC offset. Use when the user asks what time it is, what's today's date, or the current datetime — e.g. `/time`, `/time utc+7`, `/time Asia/Bangkok`, `/time America/New_York`.

0
installs
▌
Todo · thaitype bundle
Manage personal tasks and daily plans. Use when the user wants to add, view, update, check off, or delete tasks. Also use when drafting or viewing a daily plan. Tasks are stored locally in my-data/tasks.jsonl (JSONL event log). Daily plans are freeform markdown files in my-data/daily-plans/YYYY-MM-DD.md.

0
installs
▌
Money · thaitype bundle
Track personal income, expenses, and transfers between accounts. Use when the user wants to record spending, add income, check balances, view transaction history, manage categories/accounts, or anything related to personal finance tracking. Trigger on mentions of money, spending, income, expenses, budget, balance, or financial tracking.

0
installs
▌
Sync Git · thaitype bundle
Sync the current git repo with remote — pull, auto-commit, push, and resolve conflicts. Use when user says "sync", "sync repo", "push changes", "save changes", or wants to commit and push all changes. Also use when user mentions conflicts during sync.

0
installs
▌
Claude Tui · thaitype bundle
Send slash commands (like /context, /compact, /cost) to a Claude Code TUI session via tmux and capture the output. Use when the user wants to run TUI-only commands on a session programmatically.

0
installs
▌
Claude Usage · thaitype bundle
Show token usage and estimated API cost for a Claude Code session. Use when the user asks about session cost, how much a session spent, token breakdown, API usage, or wants to analyze spend over a date range.

0
installs
▌
Sandbox Init · thaitype bundle
Scaffold an isolated Docker dev sandbox for the current project directory. Generic — not tied to any specific fleet or crew concept; works for any project. Use when the user wants to set up (or a project needs) an isolated container for all builds/runs/tests, e.g. "/sandbox-init" or "/sandbox-init myproject".

0
installs
▌
Sandbox Start · thaitype
Reference for working inside a project that already has a sandbox/ directory (from sandbox-init) — the sandbox rule, exec/shell conventions, and how to think about upgrading it over time. No parameters, no automation. Read this before doing dev work in such a project, or when the user asks how to work with / upgrade an existing sandbox.

0
installs
▌
Claude Context · thaitype bundle
Check context usage and session info for any Claude Code session. Use when the user asks about context usage, token count, session info, or wants to list/inspect Claude sessions.

0
installs
▌
Claude Session Id · thaitype bundle
Find the latest Claude Code session ID(s). Use when the user asks for the current or most recent session ID, wants to find a session, or needs to look up recent sessions to resume or inspect.

0
installs
▌
Docker Playwright · thaitype
Run Playwright tests inside a Docker container (the long-lived "pw" container running the official mcr.microsoft.com/playwright image, with $HOME/gits mounted at /work). Use when the user wants to run, debug, or report Playwright tests via Docker, e.g. "run playwright in docker", "/docker-playwright tests/smoke.spec.ts", or "run the smoke tests in the pw container".

0
installs
▌
Claude Session Log · thaitype bundle
Show a full action log (text, tool calls, tool results) for a Claude Code session. Use when the user wants to replay, debug, or inspect what happened in a session, especially after running claude -p.

0
installs
▌
Typ Do Work · thaitype
Doing work assigned by a director — prove your checks can actually fail, verify from a cold state rather than your warm machine, report not-done when a check does not hold, and name what you did not establish. Use when you are the crew doing the work rather than directing it, and before reporting anything as done.

0
installs
▌
Typ Direct Work · thaitype
Directing work carried out by other crews — hold the commissioner's literal brief, follow up on a fixed timer that reads its state from a file, filter every arrival against the right engagement, verify the load-bearing claim yourself, and close only on the commissioner's quoted acceptance. Use when work is assigned to you to direct rather than to do, and again whenever anything arrives while an engagement is open.

0
installs

31 published skills