← Back to Skills Marketplace
222
Downloads
1
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install asking-until-100
Description
Repo-aware questioning protocol for OpenClaw that increases clarification before acting on coding, project-build, architecture, debugging, and implementation...
README (SKILL.md)
Asking Until 100
Overview
Use this skill to slow down execution when the task is underspecified, risky, or expensive to get wrong. Treat "100" as target readiness to proceed, not literal certainty.
Workflow
- Load explicit instructions and repo-local config such as
.asking-until-100.yaml. - Classify the task as
coding,build,architecture,debugging,discovery, orgeneral. - Inspect the repo when it looks relevant so repo-discoverable facts do not turn into avoidable questions.
- Estimate readiness from the configured dimensions in
references/protocol.md. - Choose a questioning mode:
fastfor low ambiguityguidedfor moderate ambiguitydeepfor higher ambiguity or requested rigorreportfor highest-rigor coding and build tasks with decision-critical gaps
- Ask the highest-value questions before taking action.
- Respect the execution gate:
- highest-rigor
codingandbuildtasks default to blocking clarification - other tasks default to explicit assumptions when gaps remain
- highest-rigor
Questioning Style
- Prefer structural, directional, and decision-shaping questions over generic filler.
- Use a working hypothesis when it helps the user react to a proposed path.
- Offer suggested answers when useful, but always leave a free-form path.
- Do not ask for facts that can be inspected directly from the repo.
High-Rigor Report
For highest-rigor coding or build tasks, begin with Provisional Project Structure, then emit:
Working Hypothesis, Architecture Questions, Product Questions, Constraint Questions, and
Decision-Critical Unknowns.
The working-hypothesis section must also summarize the execution gate and blocking dimensions.
See references/coding-report-format.md for the required output order and
scripts/render_project_structure.py for deterministic structure rendering.
References
references/protocol.mdfor readiness, repo-aware escalation, and stop conditionsreferences/config.mdfor config fields, precedence, and asking-intensity behaviorreferences/question-patterns.mdfor question quality rules and option patternsreferences/coding-report-format.mdfor the high-rigor report contractreferences/build-playbook.mdfor build-specific gaps to check before acting
Scripts And Assets
scripts/validate_config.pyvalidates profile filesscripts/preview_question_report.pypreviews questioning output for a promptscripts/render_project_structure.pyrenders prompt-only or repo-aware provisional structuresscripts/explain_profile_merge.pyshows the effective merged profileassets/contains bundled profiles tuned forgpt-5.4withxhighreasoning assumptions
Keep this file concise. Use the references for detailed policy, config, and output examples.
Usage Guidance
This skill appears to do what it claims: inspect the local repo and produce structured clarification questions and reports before making changes. Before installing or enabling it: (1) review any repo-local .asking-until-100.yaml you may have to ensure it doesn't contain secrets or sensitive tokens, (2) glance through the bundled scripts (render_project_structure.py, planner.py, validate_config.py) to confirm they only read local files (they do) and do not call external networks, and (3) be aware that the agent may read your workspace files when you invoke this skill — avoid storing sensitive credentials in the repo if you don't want them inspected. If the skill later requested network endpoints, environment tokens, or an install hook that downloads code, re-evaluate as those would be out-of-scope for a questioning/reporting tool.
Capability Analysis
Type: OpenClaw Skill
Name: asking-until-100
Version: 0.1.0
The 'asking-until-100' skill bundle is a meta-protocol designed to increase AI agent reliability by enforcing a clarification phase before task execution. The Python scripts (planner.py, render_project_structure.py) implement logic for repository inspection and task classification using safe practices, such as yaml.safe_load and non-recursive directory walking limited to metadata. There is no evidence of data exfiltration, network activity, or malicious execution; rather, the instructions in SKILL.md and the reference documents act as a defensive measure by requiring the agent to verify requirements and repo context before proceeding.
Capability Assessment
Purpose & Capability
The name/description (repo-aware questioning for coding/build tasks) matches the included assets and scripts: planner.py, render_project_structure.py, preview/validate scripts, and bundled profiles. There are no unrelated requirements (no cloud credentials, no system-level config paths) that would contradict the stated purpose.
Instruction Scope
SKILL.md explicitly instructs the agent to load repo-local config (e.g. .asking-until-100.yaml), inspect the repo when relevant, and generate structured question reports. The shipped scripts implement repo inspection, profile merging, structure rendering, and validation. The instructions and code reference only local repo files and bundled profile assets; they do not attempt to read unrelated system state, secrets, or external endpoints.
Install Mechanism
No install spec is provided (instruction-only), and all code is bundled with the skill. There are no downloads or external installers referenced in the manifest, so nothing arbitrary is fetched at install time.
Credentials
The skill declares no required environment variables or primary credential. The code and profiles include references to a 'secrets' readiness dimension but do not request or access environment secrets directly. The requested inputs are proportional to a repo-aware questioning tool.
Persistence & Privilege
always is false and the skill is user-invocable. It does not request persistent or elevated privileges, and there is no indication it modifies other skills or system-wide agent settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install asking-until-100 - After installation, invoke the skill by name or use
/asking-until-100 - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.1.0
- Initial release of the "asking-until-100" skill for OpenClaw.
- Introduces structured questioning protocol to clarify tasks before execution, focusing on coding, build, architecture, and debugging.
- Implements readiness estimation and adjustable questioning modes (fast, guided, deep, report) based on ambiguity and rigor.
- Enforces high-rigor question reports for critical coding/build tasks, including provisional project structure and decision-critical unknowns.
- Integrates repo awareness to avoid unnecessary questions and references supporting scripts and configuration options.
Metadata
Frequently Asked Questions
What is Asking Until 100?
Repo-aware questioning protocol for OpenClaw that increases clarification before acting on coding, project-build, architecture, debugging, and implementation... It is an AI Agent Skill for Claude Code / OpenClaw, with 222 downloads so far.
How do I install Asking Until 100?
Run "/install asking-until-100" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Asking Until 100 free?
Yes, Asking Until 100 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Asking Until 100 support?
Asking Until 100 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Asking Until 100?
It is built and maintained by Hongyi3 (@hongyi3); the current version is v0.1.0.
More Skills