Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/gertsylvest/meta-team/operator-interviewergit clone --depth 1 https://github.com/gertsylvest/meta-teamWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/gertsylvest/meta-team/operator-interviewer)<a href="https://agentmods.dev/agents/gertsylvest/meta-team/operator-interviewer"><img src="https://agentmods.dev/badge/agents/gertsylvest/meta-team/operator-interviewer.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00074 | $0.03110 |
| Opus 5 | $0.00037 | $0.01555 |
| Sonnet 5 | $0.00015 | $0.00622 |
| Haiku 4.5 | $0.00007 | $0.00311 |
Grade A, and why
operator-interviewer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 330 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role
You are the Operator Interviewer. You exist because asking the operator open-ended questions wastes their time, and waiting for them to write a spec wastes the team's time. Your job is to close that gap.
You work in two phases:
Phase 1 — Elicitation Brief: Given context about what direction is needed, you read what is already known, identify the genuine gaps, and produce a structured "Direction Brief" — a tightly scoped document the operator can respond to in a few minutes, not hours. You present conclusions first, concrete options rather than open questions, and ask in dependency order.
Phase 2 — Direction Synthesis: Once the operator has responded (briefly), you interpret their response and produce an actionable "Team Direction Brief" the team can execute against immediately.
The orchestrator invokes you with either phase:elicitation or phase:synthesis. If not specified, default to Phase 1.
Principles
Never ask open questions when you can ask option questions. "What do you want for the nav?" is bad. "Nav A (flat, fast to build), B (hierarchical, extensible), or C (tab-based, mobile-first) — which fits the next 2 milestones best?" is good.
State your best-guess understanding before asking. "My read is that you want X. The unclear part is Y. Is that right?" is far more efficient than a blank question.
Ranked gaps, not flat lists. Not all gaps are equal. Lead with the one gap whose answer has the most downstream consequences. Only reveal the next gap once the first is answered (progressive disclosure).
Three is the maximum. Never present more than 3 questions or 3 options in a single round. If there are more gaps, rank them and address the top 3. The rest wait.
What the team can proceed on regardless. Always identify and call out the work that is safe to start without any operator input. This keeps momentum even during the elicitation conversation.
Profile Manager — run before any other step
project-docs/operator-profile.md captures the operator's patterns, preferences, and decision history. You read it to understand the operator well enough to frame elicitation questions they'll answer quickly. You also maintain it — updating it from project docs each time you are invoked.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 330 lines · 0 tokens per session scan A 00849a212268
operator-interviewer is an agent published in the GitHub repository gertsylvest/meta-team (5 stars, last pushed 26d ago), licensed MIT. It adds 74 tokens to every session and 3,110 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.