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 OutlineDriven/outline-driven-development --skill batch-ask-megit clone --depth 1 https://github.com/OutlineDriven/outline-driven-developmentWrote 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/outlinedriven/outline-driven-development/batch-ask-me)<a href="https://agentmods.dev/skills/outlinedriven/outline-driven-development/batch-ask-me"><img src="https://agentmods.dev/badge/skills/outlinedriven/outline-driven-development/batch-ask-me.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
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 →
- medium Excessive Agency · line 29 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 35 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 | $0.00056 | $0.01048 |
| Opus 5 | $0.00028 | $0.00524 |
| Sonnet 5 | $0.00011 | $0.00210 |
| Haiku 4.5 | $0.00006 | $0.00105 |
Grade A, and why
batch-ask-me 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 — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Batch ask me
Contract
| Field | Bound contract |
|---|---|
| Trigger | The user faces multi-fork decisions, unresolved prerequisites, or explicitly says "batch ask me" or "clarify the design space". |
| Authority | Read-only: no file, VCS, credential, paid, published, deployed, or remote mutation. |
| Side effect | Conversation-only question batches and decision-tree state. |
| Done | Frontier empty, all branches of the design tree visited, and shared understanding confirmed by the user. |
Inputs
The user's problem or design space to explore, stated in the conversation or inferable from it. Optional: any decisions already settled by prior context, which seed the design tree as resolved nodes.
Procedure
- Run Verbalized Sampling before round one. Sample multiple intent hypotheses, each with an explicit weight on a 0 to 1 scale and a concrete falsifier. Present the weighted hypotheses and falsifiers visibly, immediately before the first question batch. Seed the design-tree roots and initial frontier from the surviving hypotheses. Done when: the weighted hypotheses and falsifiers are presented and the design-tree roots and frontier are seeded.
- Build a design tree where each node is a decision. The frontier is every decision whose prerequisites are already settled — the questions to ask now without guessing at answers not yet heard. Done when: the design tree is built and the frontier is computed.
- Each round, ask the whole frontier as a batch of single-select questions, each with a recommended answer. Wait for the user's answers, then recompute the frontier and ask the next round. A question whose answer depends on another open question belongs to a later round, not this one. Done when: the frontier batch is asked and answers are collected, or the frontier is empty.
- Question shape: one single-select question per axis, mark the recommended option first with "(Recommended)", at most four questions per fire, and never use multiSelect for override semantics. Done when: every question follows the single-select shape with a marked recommendation.
- When the frontier contains more than four questions, keep the whole frontier in one round. Route the four highest-impact questions through the question tool; put every remaining question in the same message as numbered Markdown in the form
**Q<n>: <question title>**followed by the body, choices, and-> Recommended: <answer>. Answers from the tool and the Markdown questions settle together in one round-trip. Recompute the frontier once after the full answer set instead of advancing four questions at a time. Done when: the full frontier is asked in one round with tool and Markdown questions settling together. - Choose the four tool questions by how many downstream decisions each answer unblocks. Break ties toward the question whose default is least safe to assume. Done when: the four highest-impact tool questions are selected.
- Finding facts is the agent's job, not the user's. When a frontier question needs an environmental fact (filesystem, tools, codebase), dispatch a sub-agent to find it; never ask the user for something that can be looked up directly. A running exploration is an unsettled prerequisite; only its downstream questions wait, so ask the rest of the frontier now. Done when: environmental facts are dispatched to sub-agents and the rest of the frontier is asked.
- Do not resample Verbalized Sampling on subsequent rounds unless user answers materially change the survivor set. If resampling is triggered, update the survivor set, adjust the design-tree roots, and recompute the frontier. Done when: resampling is skipped or triggered with the survivor set and frontier updated.
- The session is done when the frontier is empty: every branch of the design tree visited, nothing left silently assumed. Do not act on the result until the user confirms a shared understanding has been reached. Done when: the frontier is empty and the user confirms shared understanding.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 41 lines · 56 tokens per session scan A f9f7c7ed7a77
batch-ask-me is a skill published in the GitHub repository OutlineDriven/outline-driven-development (52 stars, last pushed 2d ago), licensed Apache-2.0. It adds 56 tokens to every session and 1,048 once invoked, about $0.0003 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 2026-09-03.
Other skills, from other repositories
askme
Explore intent with Verbalized Sampling before planning. Use when the task is ambiguous, you need maximum clarifying questions, or you want collaborative or adversarial mode via /askme.
slicing-code-context
Use when an exact symbol, path, entrypoint, or line range can bound a focused code question or patch proposal under a fixed source budget. Not for source changes or broad repository exploration.
handoff-prompt
Use when the user asks for a handoff, delegation, or clipboard-ready prompt for another agent: a standalone path-free prompt copied to the clipboard, confirmed by title. Not for session-snapshot briefs — use handoff; never remote, credential, publish, deploy, or irreversible.
autopilot
Run a hands-off plan-to-ship pipeline by chaining existing skills. Use when the user says "autopilot", "take this from plan to shipped", "run the whole pipeline", "hands-off ship it", or "do the end-to-end build".
llm-self-loop
Use when a button click, dashboard check, or human verdict sits inside an iteration loop. Replaces it with an autonomous gate or moves non-automatable work outside the loop. Don't use for remote, credential, publish, deploy, or irreversible changes.
catchup
Use when the human returns after a gap, cannot follow the project, asks what happened, or wants a visual HTML recap page. Not for onboarding: use onboard. Not for handoff: use handoff.