メニュー
Move issues through a triage state machine. Use when reviewing bugs, feature requests, or preparing issues for implementation.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Find deepening opportunities in the codebase — refactors that reduce coupling, simplify interfaces, and improve AI-navigability. Use when improving architecture, finding refactoring opportunities, or making a codebase easier to work with.
Use before any feature work — explores intent, asks clarifying questions one at a time, produces a design doc. Invoke before writing any code or plan.
Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "less tokens", or invokes /caveman.
Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → review. Use when something is broken, throwing errors, or regressing.
Interview the user or a design relentlessly until reaching shared understanding. Walk down every branch of the decision tree, resolving dependencies one by one. Use when stress-testing a plan, clarifying design, or the user says "grill me".
Use at the start of any session to understand which skill to invoke. Maps the full workflow — feature development, bug fixing, architecture, and utilities — for a C++ game server project with SVN.
| name | triage |
| description | Move issues through a triage state machine. Use when reviewing bugs, feature requests, or preparing issues for implementation. |
Move issues through a state machine of triage roles.
Every comment posted during triage must start with:
> *This was generated by AI during triage.*
Category:
bug — something is brokenenhancement — new feature or improvementState:
needs-triage — needs evaluationneeds-info — waiting on reporterready-for-agent — fully specified, AFK-ready (can be implemented without human input)ready-for-human — requires human judgment or external accesswontfix — will not be actionedEvery triaged issue should carry one category and one state role.
Invoke /triage and describe what you want in natural language:
Present three buckets (oldest first):
needs-triage — evaluation in progressneeds-info with reporter activity — needs re-evaluationShow counts and a one-line summary per issue.
Gather context. Read the full issue. Explore the relevant codebase using domain vocabulary.
Recommend. State your category and state recommendation with reasoning. Wait for direction.
Reproduce (bugs only). Trace the relevant code path. Report whether the repro steps are coherent and what the likely code path is. A confirmed code-path trace makes a much stronger agent brief.
Apply the outcome:
ready-for-agent → post an agent brief (see template below)ready-for-human → same structure, note why it can't be delegatedneeds-info → post triage notes (see template below)wontfix → polite explanation and close> *This was generated by AI during triage.*
## What to Build
[Concise description of the fix or feature, using domain vocabulary]
## Context
[Relevant code paths, module names, existing patterns to follow]
## Acceptance Criteria
- [ ] ./qmake.sh compiles clean
- [ ] Code review passed
- [ ] svn commit done with descriptive message
## C++ Conventions Reminder
- `UtilArith::safeSub()` for subtraction
- `UtilSTL::findMapPtr` for map lookups
- `setChanged()` after data modifications
- `sendResponse()` only in CmdParser layer
- Full braces for all control flow
> *This was generated by AI during triage.*
## Triage Notes
**What we've established:**
- [point 1]
**What we need from you (@reporter):**
- [specific question 1]
- [specific question 2]