Runs grounded question-and-answer sessions against supplied context and saves session summaries.
npx skills add https://github.com/Arasz/ai-badger --skill qa-arasz
Please help me install this Agent Skill. Skill: qa Source: https://github.com/Arasz/ai-badger/tree/main/features/common/skills/qa Command: npx skills add https://github.com/Arasz/ai-badger --skill qa-arasz
docs/qa-sessions/<topic-slug>.md. - Use Case: Paste an architecture RFC into the chat, ask a series of clarifying questions about its trade-offs, receive answers each citing the RFC sections and codebase files they rest on, then end the session and get a durable summary file committed to the repo. ## Quick Start Start a qa session against the attached design document and answer my questions with cited sources, then save the session summary.High-intent search queries and answers about installing and using this skill.
Invoke the qa skill with the context, such as a pasted text, file path, directory, or topic. The skill restates what the session stands on, then answers each question against that context with cited sources before saving a summary.
Every answer draws on at least two independent sources, with a mandatory memory-or-code leg from project memory, the knowledge graph, or codebase search, plus a web leg when available. Missing web sources produce an explicit fallback note rather than silence or fabrication.
The skill asks for the context first and does not answer until it exists. A session with nothing to stand on would produce ungrounded answers, so no context means no answers.
Summaries are written to docs/qa-sessions/<topic-slug>.md in the consumer repo, with the directory created on demand. The slug is lowercased, hyphenated, and truncated to about 64 characters, and re-running a session overwrites the existing file.
Answers are scoped strictly to the supplied context, so questions drifting beyond it are flagged as out of scope rather than answered freely. The skill also refuses adjacent work like refactoring or task creation during the session.
khapn-2933Scan codebases with parallel agents for file discovery and context gathering.
microclawConvert raw meeting notes and transcripts into structured minutes with decisions and action items.
AraszRuns grounded question-and-answer sessions against supplied context and saves session summaries.
brainbrewlabsScout codebases by spawning parallel Explore subagents with 3-minute timeouts.