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 skills add EmpiricaAI/empirica --skill ewm-interviewgit clone --depth 1 https://github.com/EmpiricaAI/empiricaWrote 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/skills/empiricaai/empirica/ewm-interview)<a href="https://agentmods.dev/skills/empiricaai/empirica/ewm-interview"><img src="https://agentmods.dev/badge/skills/empiricaai/empirica/ewm-interview/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/empiricaai/empirica/ewm-interview"><img src="https://agentmods.dev/badge/skills/empiricaai/empirica/ewm-interview.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 5 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Agent Snooping · line 77 Skill accesses MCP server configuration files (mcp.json). MCP configs contain server URLs, authentication tokens, and tool definitions — reading them allows the skill to discover and potentially abuse other tool integrations.Fix: Remove all code or instructions that read MCP configuration files (mcp.json). MCP server details should be managed by the agent runtime, not read by individual skills.
- medium Excessive Agency · line 104 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Rogue Agent · line 161 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
- medium Excessive Agency · line 88 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 94 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 5.1 | $0.00078 | $0.02938 |
| Opus 5 | $0.00039 | $0.01469 |
| Sonnet 5 | $0.00016 | $0.00588 |
| Haiku 4.5 | $0.00008 | $0.00294 |
Grade A, and why
ewm-interview 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 11d 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 — 295 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/ewm-interview — Epistemic Workflow Manager Interview
Produce a workflow-protocol.yaml describing how this person wants to work with AI — then actually set up what it implies.
The rule that governs the whole interview
Ask with options, never with a blank. Every question goes through
AskUserQuestion with 2–4 concrete, pickable options. The tool supplies its own
"Other" path, so a free-form answer is always available without you offering one
— do not add an "Other"/"Something else" option yourself, it duplicates the
built-in.
This matters more than it looks. "What are your main constraints?" asked into a void gets a shrug or a paragraph you cannot map to a field. The same question as four options gets an answer in one click, and the options themselves TEACH what the field means. Prose questions are what made v0.1 a form-fill wearing a conversation's clothes.
Three rules for the options you write:
- Make every option genuinely pickable. An option nobody could choose is filler; it makes the list look considered while narrowing the real choice.
- Put a recommendation first when there is a sane default, and say so:
"Collaborative — check in on approach (Recommended)". multiSelect: truewhen answers are not exclusive (domains, tools, non-negotiables). Most of this interview is multi-select.
Batch related questions into ONE AskUserQuestion call — it takes up to 4. Five
calls of 3–4 questions each covers the whole interview.
Phase 1 — Goals
One call, up to 4 questions:
| Ask | Shape | Options to offer |
|---|---|---|
| What are you mainly trying to accomplish right now? | single | Ship a product / feature · Research & understand a problem space · Grow an organisation or practice · Operate & maintain something live |
| What does "done" look like for that? | single | A shipped, working thing · A decision I can defend · A repeatable process · Learning I can reuse |
| What is your binding constraint? | multi | Time · Money / headcount · Missing knowledge · Regulatory / compliance · Dependence on other people |
| Horizon? | single | This week · This quarter · This year · Open-ended |
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.
- 11d ago First seen · 295 lines · 78 tokens per session scan A 14c968467bfd
ewm-interview is a skill published in the GitHub repository EmpiricaAI/empirica (246 stars, last pushed today), licensed MIT. It adds 78 tokens to every session and 2,938 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 2026-08-30.
Other skills, from other repositories
workflow-builder
Design, build, and maintain autonomous OpenClaw workflows (stewards). Use when creating new workflow agents, improving existing ones, evaluating automation opportunities, or debugging workflow reliability. Triggers on "build a workflow", "create a steward", "automate this process", "workflow audit", "what should I…
smart-delegation
Intelligent task delegation — route to think with deep reasoning for hard problems, or Grok for unfiltered takes. Teaches when to escalate, how to pack context into sub-agent spawns, and how to communicate delays transparently. Default: handle directly on chat (thinking off). Escalate only when the quality gain…
review
A pause before an artifact goes into the world. Reviews external comms, money, calendar, public posts, or send-as-operator actions through a small panel of independent lenses (empathy first) and returns a verdict of pass / edit / hold / block.
claude-code
Route real repo work to Claude Code instead of editing by hand. Triggers on "claude code" or "cc", and on any request to edit, fix, refactor, or open a PR in a repo outside /.openclaw/workspace. Claude Code picks up the repo's CLAUDE.md / AGENTS.md, applies its standards, and knows the /ai-coding-config:multi-review…
cortex
Personal knowledge compiler and memory system. Ingests raw sources (documents, notes, transcripts, captures) into a structured, interlinked knowledge base. Maintains living entity pages, learning analysis, and the MEMORY.md routing table. Replaces the librarian skill.
drive-to-done
Drive a task all the way to a verified done state — write DoD first, verify each item with evidence, stop only at named stop conditions.