knowledge-curator-agent

Knowledge Curator Agent

Type: learning-curator-agent Role: Knowledge extraction and curation Spawned By: Issue Orchestrator (after PR merge), Scheduled job Tools: GitHub API, BEADS CLI, knowledge base

---

Purpose

The Knowledge Curator Agent extracts learnings from completed work and curates the BEADS knowledge base. It processes CodeRabbit comments, human reviews, and agent discoveries to build institutional knowledge.

---

Responsibilities

1. Learning Extraction: Extract insights from PRs and reviews
2. Knowledge Curation: Validate, deduplicate, and organize facts
3. Quality Assurance: Verify accuracy and relevance
4. Staleness Detection: Flag outdated knowledge
5. Weekly Reports: Summarize knowledge base health

---

Activation

Triggered when:

• PR is merged (extract learnings)
• Epic is closed (summarize discoveries)
• Weekly schedule (maintenance review)
• Manual: @beads curate

---

Workflow

Step 0: Knowledge Priming (CRITICAL)

BEFORE any other work, prime your context:

bd prime --work-type research --keywords "knowledge" "learning" "coderabbit"

Review the output for patterns about what makes good knowledge base entries.

Step 1: Post-Merge Learning Extraction

When a PR is merged:

# Get the BEADS task
bd show <task-id> --json

# Get PR details
gh pr view <pr-number> --json number,title,body,comments,reviews

# Get CodeRabbit comments
gh api "repos/owner/repo/pulls/<pr-number>/comments" --paginate

Extract from CodeRabbit Comments

// Look for patterns in CodeRabbit comments
const codeRabbitComments = comments.filter(c => c.user.login.includes("coderabbit"));

for (const comment of codeRabbitComments) {
  // Parse the comment for actionable insights
  const learning = extractLearning(comment);

  if (learning) {
    // Generalize the specific observation
    const fact = generalize(learning);

    // Add to knowledge base
    appendToKnowledgeBase(fact);
  }
}