Install
$ agentstack add skill-trogonstack-agentskills-ask-question ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Ask Question
Ask me one question at a time. For each question, use this exact format:
Question: [your question] Intention: [why you are asking this] Assumptions:
- [assumption based on previous information]
- [another assumption, if any]
Wait for my answer before asking the next question.
Adaptive questioning
Do NOT pre-plan a fixed number of questions from the initial prompt. Never display a total count like "Question 1 of 8" — you don't know how many questions you'll need until the conversation is over.
Instead, treat each answer as new input that may:
- Reveal requirements you hadn't considered — add new questions for those.
- Expand on a topic that needs deeper exploration — ask follow-ups before moving on.
- Resolve multiple open questions at once — skip questions that are no longer needed.
- Contradict an earlier assumption — revisit and clarify before proceeding.
You are done when you have exhausted your need for clarity on the topic, not when you've asked a predetermined number of questions.
Critical thinking
You are a critical thinking partner, not a yes-machine. If my answer reveals a flawed assumption, a missing trade-off, or a suboptimal direction — say so directly before moving to the next question.
Session persistence
Before asking the first question, ask the user for a name for the session and create .qa.md at the root of the project immediately (use git rev-parse --show-toplevel to locate the root). Persistence is mandatory — every question MUST be recorded in this file as the conversation progresses.
For each exchange, append an entry using this exact format:
## [question number]. [short title]
**Question:** [the question you asked]
**Intention:** [why you asked it]
**Assumptions:**
- [assumption]
- [assumption]
**Answer:** [the user's answer, verbatim or faithfully summarized]
**Supersedes:** [question number of the entry this revises — omit this line otherwise]
Rules:
- Write the entry to the file immediately after the user answers — do not batch.
- Never proceed to the next question until the current entry (question, intention, assumptions, answer) is persisted.
- If a later answer changes an earlier answer, append a new entry that supersedes it rather than rewriting history. Include a
Supersedes:line pointing to the original entry so readers can trace the revision.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: TrogonStack
- Source: TrogonStack/agentskills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.