agentleFS
Sign inStart free
anthonysuherlianthonysuherli/br8n/skills/explore/SKILL.md

explore

Skill1 starsChanged 3 months agoMITView on GitHub

Force the br8n gap-fill pipeline — plan → web search → crawl → extract → merge — to seed or deepen the repo+branch KB from the web, then rebuild the synopsis. Use when /br8n:pickup returns coverage='gap', when starting a KB from scratch, or when the user explicitly wants fresh online research persisted into the session KB.

We split this skill into the rules it gives an agent

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 50 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run the exploration pipeline to grow the KB from the web.

    line 8

“Common” is measured: kept by at least 6 in 10 of the 50 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

br8n — Explore

Run the exploration pipeline to grow the KB from the web. This is the grow step that /br8n:search calls conditionally — invoke it directly when you already know the KB is thin and want to fill it.

Step 0 — Resolve target + tap

Follow ../_shared/preamble-first.md: resolve project/kb from git, then mcp__plugin_br8n_br8n__br8n_resume(project, kb, query=<the topic>) to confirm the gap is real and to target the prompt at the thin areas rather than re-fetching what's already known. Show the resume card.

Step 1 — Shape the prompt

Turn the user's topic into a focused exploration prompt. If the resume card showed partial coverage, narrow the prompt to the missing slice (don't re-explore covered ground). Pass max_findings only if the user wants to cap breadth; otherwise let the config default apply.

Step 2 — Run (blocking)

mcp__plugin_br8n_br8n__br8n_explore(project, kb, prompt=<focused prompt>, max_findings?)

This blocks 1–3 minutes (plan→search→crawl→extract→merge). Tell the user it's running before you call it. On return you get {finding_count, finding_ids}; the synopsis rebuild is scheduled automatically.

Step 3 — Report + re-ground

State how many findings were persisted. Then re-tap with mcp__plugin_br8n_br8n__br8n_resume(project, kb, query=<topic>) to show the now-richer card and answer whatever prompted the explore. The next /br8n:pickup will reflect the new knowledge.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free