kb-advisor
Skill12 starsChanged 4 months agoMITView on GitHub
Knowledge base advisor — search, synthesize, and update your team's knowledge repositories.
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
Search and synthesize answers from your team's knowledge base repositories.
- Distinctive: line 63
Read the most relevant files (top 3-5 matches)
- Distinctive: line 70
When the conversation produces novel insights worth preserving:
- Distinctive: line 82
Always cite the specific source file
- Distinctive: line 83
If the knowledge base doesn't cover a topic, say so — don't fabricate
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Knowledge Base Advisor
Search and synthesize answers from your team's knowledge base repositories.
Configuration
Knowledge bases are configured during superskills setup and stored in ~/.superskills/knowledge.conf.
# Check configured knowledge bases
cat ~/.superskills/knowledge.confEach line is: name|path|description
Example:
docs|~/.superskills/knowledge/docs|Internal documentation and architecture decisions
research|~/.superskills/knowledge/research|Market research and competitive analysisWorkflow
1. Discover Available Knowledge
# List configured knowledge bases
cat ~/.superskills/knowledge.conf | while IFS='|' read -r name path desc; do
path="${path/#\~/$HOME}"
if [ -d "$path" ]; then
count=$(find "$path" -name "*.md" -o -name "*.txt" -o -name "*.html" | wc -l | tr -d ' ')
echo " [$name] $count files — $desc"
else
echo " [$name] NOT CLONED — $path"
fi
done2. Search Across Knowledge Bases
# Search all knowledge bases for a topic
for kb_path in $(awk -F'|' '{print $2}' ~/.superskills/knowledge.conf); do
kb_path="${kb_path/#\~/$HOME}"
[ -d "$kb_path" ] && grep -rli "search terms" "$kb_path" --include="*.md" --include="*.txt" --include="*.html"
done3. Synthesize Advice
- Read the most relevant files (top 3-5 matches)
- Quote specific passages with attribution (file name and path)
- Cross-reference multiple sources when they agree or disagree
- Connect abstract principles to the user's specific situation
4. Update Knowledge Base (Optional)
When the conversation produces novel insights worth preserving:
# Add new knowledge to the appropriate repo
echo "content" > <kb-path>/new-insight.md
# Commit the update
cd <kb-path> && git add -A && git commit -m "Add: <topic>"Tips
- Always cite the specific source file
- If the knowledge base doesn't cover a topic, say so — don't fabricate
- Suggest pulling latest:
cd <kb-path> && git pull - For questions outside the knowledge base, answer from general knowledge but note the gap
- Multiple knowledge bases can be searched together for cross-domain insights
In this repository42 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- a11ySkill
- articleSkill
- basic-reviewSkill
- cache-strategySkill
- clean-codeSkill
- cli-for-agentSkill
- content-writerSkill
- daily-qaSkill
- dbmap-auto-offSkill
- dbmapSkill
- db-optimizeSkill
- debugSkill
- defenseSkill
- diff-walkthroughSkill
- evalSkill
- executing-plansSkill
- finish-branchSkill
- fuzzSkill
- handoffSkill
- humanizeSkill
- iac-scanSkill
- minimalist-reviewSkill
- pentestSkill
- perf-profileSkill
- playwrightSkill
- pricingSkill
- qa-fullSkill
- quota-resilienceSkill
- subagent-driven-developmentSkill
- superskills-doctorSkill
- superskills-upgradeSkill
- tapestrySkill
- tddSkill
- test-coverageSkill
- validate-ideaSkill
- verifySkill
- web-perfSkill
- worktreesSkill
- write-planSkill
- youtubeSkill
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.

