Exécutez n'importe quel Skill dans Manus
en un clic

Exécutez n'importe quel Skill dans Manus en un clic

interrogation

Étoiles7

Forks0

Mis à jour20 mai 2026 à 19:51

Use when the user invokes `/interrogation "<prompt>"`. Ask focused clarifying questions via AskUserQuestion until you have enough certainty to act on the prompt without guessing. Stop early — three rounds at most.

Installation

Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.

Exécuter dans Manus

Source

himattm

himattm/skills

Ouvrir le dépôt GitHub Voir les dépôts du créateur

Téléchargement

Exécuter dans Manus

Métiers associésSOC

Basé sur la classification professionnelle SOC

Représentants du service clientèleBureau et soutien administratif·SOC 43-4051

SKILL.md

readonly

name	interrogation
description	Use when the user invokes `/interrogation "<prompt>"`. Ask focused clarifying questions via AskUserQuestion until you have enough certainty to act on the prompt without guessing. Stop early — three rounds at most.

Interrogation

The user gave you a prompt that's likely under-specified. Drill in with structured questions before doing the work, so the output lands on the first try instead of needing a redo.

Workflow

Read the prompt carefully. Identify the 2–4 ambiguities that, if guessed wrong, would lead to materially different output. These are the only things worth asking about.
Ask via AskUserQuestion. Batch related questions into one call — never one question per turn. Use multiple-choice when the answer space is bounded; use a free-form option for open ends.
Outline your understanding after every answer. Don't silently decide "I have enough." Write a short, high-level outline of what you now think the user wants — scope, output shape, key choices — so the user can see your mental model and correct it before more time is invested. Bullet points are fine. Keep it to ~5 lines.
Iterate at most twice more. If the outline reveals a remaining ambiguity, ask another batched round. If the user confirms the outline (or says "go" / "ship it"), stop. Three rounds maximum.
Final confirm and proceed. Once the outline matches, restate the agreed plan in one paragraph and begin the work.

What to ask about

Scope — what's in, what's out
Output shape — file? PR? branch? message? what does "done" look like?
Hard constraints — must use X, can't touch Y, deadline, audience
Judgement calls the user has a view on — speed vs. polish, convention vs. novelty

What NOT to ask

Anything answerable from the codebase, recent context, or CLAUDE.md
Stylistic micro-decisions — pick the codebase convention silently
Things the user already answered earlier in the session
Yes/no when a multiple-choice would be more decisive

Stop conditions

The user confirms your outline (mental models match)
The user says "go", "ship it", "you have enough", or equivalent
Three rounds in, further questions would scope-creep rather than clarify

Plus depuis ce dépôt

même dépôt

address-review

himattm/skills

Use when the user asks to address PR review comments — from human reviewers or automated bots (Gemini, Coderabbit, Copilot, SonarCloud, etc.) — to triage and respond to feedback on a pull request

2026-05-207

android-cli

himattm/skills

Orchestrates Android development tasks including project creation, deployment, SDK management, and environment diagnostics using the `android` command-line tool.

2026-05-207

android-coroutine-trace

himattm/skills

Use to find stuck coroutines, leaked jobs, and suspended awaits with no resumer — the concurrency bugs that don't crash but cause hangs, missing UI updates, or memory creep. Ephemerally add `kotlinx-coroutines-debug`, install `DebugProbes` in `Application.onCreate`, run the suspect flow, dump active coroutines with stack traces, fix, then remove the dependency and the install/dump calls. Reach for this when `Log.d` shows a `launch` ran but `collect` never received, or a screen leaves and something keeps running.

2026-05-207

android-crash-repro-loop

himattm/skills

Use to flush out intermittent crashes and ANRs ("it crashes sometimes," "fails on slow devices," "happens after rotation a few times") by scripting the trigger sequence with `adb shell input` and running it in a shell loop against a cleared logcat, stopping on first match for `FATAL EXCEPTION` or `ANR in`. Inject deterministic stress (rotation, slow network, low memory) to surface latent races. The empirical answer to "is this still a bug, and under what conditions?"

2026-05-207

android-perfetto-analyze

himattm/skills

Use to extract concrete numbers and slice-level evidence from a `.perfetto-trace` captured by `android-perfetto-capture`. Run SQL queries against the trace via the `trace_processor` binary, then hand the (small) result file to a Sonnet sub-agent for verdict — never read raw trace data inline. This is where "the app feels slow" becomes "frame 42 missed budget by 8.3 ms because `MainActivity.onResume` ran 18 ms on `main` while the IO dispatcher was idle." The analysis half of the perfetto loop.

2026-05-207

android-perfetto-capture

himattm/skills

Use to capture Perfetto system traces from a connected Android device — the input for any timing, threading, jank, or "is this on the main thread?" investigation. Writes a self-contained `.perfetto-trace` file to `/tmp` that downstream skills (`android-perfetto-analyze`, `android-trace-sections`) read with the `trace_processor` SQL backend. Pick the strategy that matches the question (one-shot, repeated, on-the-fly), drive the suspect flow during the recording window, and verify the trace contains the expected slices before declaring capture done.

2026-05-207

Interrogation

The user gave you a prompt that's likely under-specified. Drill in with structured questions before doing the work, so the output lands on the first try instead of needing a redo.

Workflow

Read the prompt carefully. Identify the 2–4 ambiguities that, if guessed wrong, would lead to materially different output. These are the only things worth asking about.

Ask via AskUserQuestion. Batch related questions into one call — never one question per turn. Use multiple-choice when the answer space is bounded; use a free-form option for open ends.

Outline your understanding after every answer. Don't silently decide "I have enough." Write a short, high-level outline of what you now think the user wants — scope, output shape, key choices — so the user can see your mental model and correct it before more time is invested. Bullet points are fine. Keep it to ~5 lines.

Iterate at most twice more. If the outline reveals a remaining ambiguity, ask another batched round. If the user confirms the outline (or says "go" / "ship it"), stop. Three rounds maximum.

Final confirm and proceed. Once the outline matches, restate the agreed plan in one paragraph and begin the work.

What to ask about

Scope — what's in, what's out

Output shape — file? PR? branch? message? what does "done" look like?

Hard constraints — must use X, can't touch Y, deadline, audience

Judgement calls the user has a view on — speed vs. polish, convention vs. novelty

What NOT to ask

Anything answerable from the codebase, recent context, or CLAUDE.md

Stylistic micro-decisions — pick the codebase convention silently

Things the user already answered earlier in the session

Yes/no when a multiple-choice would be more decisive

Stop conditions

The user confirms your outline (mental models match)

The user says "go", "ship it", "you have enough", or equivalent

Three rounds in, further questions would scope-creep rather than clarify