problem-brief
Publishes and maintains a problem brief: a page of findings, open decisions and questions for the operator, each with the problem, evidence and a recommended way out, kept as JSON that records every ruling so nothing is asked twice. Use by default, without being asked, whenever work turns up something the operator must decide (a finding with more than one way out, a risk, an open question, a choice between options), even mid-task and even when the operator only asked for an investigation. Also use when asked to build, publish, refresh or update a brief, findings report, open-questions or decisions page; when the operator answers, rules on or comments on anything a brief holds, in chat or in a comment on the page; when questions would otherwise go out through AskUserQuestion; and when retiring a brief whose work is done. Not for a progress, status or completion report with nothing to decide, for running a chorus review, or for converting chorus records to the shared schema.
Target: Target not categorized. Automatically collected · unverified.
Excerpt from the author’s documentation
Extracted from a fixed version of SKILL.md in its original language. This is the author’s documentation; installation, compatibility and runtime results have not been verified by this site.
Read the original English text
Topics in the source
- Problem Brief
- Overview
- When to use
- Called as /problem-brief with nothing else asked — walk the open questions
- When the operator answers or rules — this turn, before anything else
- Before raising anything — search, by class
- Start of every run — state check
- The entry — five parts, in this order
A problem brief is a published page whose one job is to let the operator rule without opening the repository. Every entry states a problem, proves it, and ends with a recommended way out. It is not a status report. The brief's JSON is the memory, not the conversation. Answers given in chat are lost when the session is summarised; to the operator, being asked again what they already answered is the worst failure this skill can have. The reader was not in your session and will not open a file. An entry that cannot be understood from the page alone is not finished. Every write to a brief's JSON goes through brief, short here for node /.claude/skills/problem-brief/bin/brief.mjs. Never edit the file by hand or with a script: each command validates the result and refuses to write anything broken, writes the change note, keeps the stamps in order and locks the file while it writes. brief --help lists every command and flag; brief set --field value changes any field the others do not cover. By default, unasked: anything the operator must rule on — a finding with more than one way out, an open decision, a question — goes into the brief for that subject (or a new one). Chat carries only the
This is an excerpt. Refer to the complete source and its current requirements.
- Source repository
- zaytsevand/problem-brief
- Source file
- SKILL.md
- Fixed revision
- 8a64f5f05306f14b15e79d28965adbd0c324e7d2
Requirements
- Only the author’s Skill metadata is included. Read the full instructions to check dependencies, licensing and conditions of use.
A small first task
先阅读原作者说明,再决定是否在自己的 AI 工具中使用。
Instructions for your AI tool
- Claude Desktop: resource-specific guide not available yet
- Claude Code: resource-specific guide not available yet
- Cursor: resource-specific guide not available yet
- Codex: resource-specific guide not available yet
Author documentation is not live verification by this site. Without evidence, we do not mark a resource as compatible, safe or free.
Sources and evidence
This record was collected automatically. Its documentation, suitability and runtime behavior have not been manually verified.
Author: zaytsevand/problem-brief · original repository metadata. Documentation review: Not reviewed. Last source check: 2026-11-09T18:01:41.484Z.
Categories are inferred from names, descriptions and paths. They do not establish compatibility.
License and attribution
MIT. Third-party content and copyright remain with their authors. Refer to the original license and terms.
Original license ↗Synchronization, copying and user-reported success do not mark a resource as runtime-tested.
Limitations and notes
- Collecting metadata does not mean this site has installed, adapted, runtime-tested or security-reviewed the resource.