continue

Continue Session

Write the next N incomplete sections (default: 3).

Context Files

• Persona: @persona.md
• Project Context: @project-context.md

Current State

• Incomplete tasks: !grep -c "\[ \]" task.md 2>/dev/null || echo "No task.md found"

Process

1. Parse Arguments

CRITICAL: Extract count, skip_review, and use_glm_tools flags from $ARGUMENTS.

skip_review = $ARGUMENTS.includes("--skip-review")
use_glm_tools = $ARGUMENTS.includes("--glm") ? "true" : "false"
count = parseInt($ARGUMENTS.replace("--skip-review", "").replace("--glm", "").trim()) || 3

The use_glm_tools value MUST be the string "true" or "false" for XML.

2. Read task.md and Extract Domain

Read task.md and identify all incomplete sections ([ ] items).

Read persona.md and extract domain information:

• Domain (from Domain Guidelines section header, e.g., "Domain Guidelines: Technology" → technology)
• If no Domain Guidelines section, default to "technology" for backward compatibility

Domain values: technology, history, science, arts, general

3. Build Section Queue

Create a queue of sections to write:

1. Collect first count incomplete ([ ]) items from task.md
2. Extract section IDs and titles for each queued item

3.5 Directory Resolution Helper

CRITICAL: Use multi-tier search to find existing research directories. This handles naming inconsistencies.

⚠️ Glob Returns Files Only

Glob does NOT return directories. All patterns MUST end with /research.md.

Pattern	Result
.research/sections/*9-1*	❌ Empty
.research/sections/*9-1*/research.md	✅ Works

Define a helper function to resolve research directories:

function resolveResearchDirectory(chapter, section, title):
  # Normalize identifiers (from research-storage skill)
  canonical_chapter = normalizeChapter(chapter)
    # "1" → "01", "01" → "01", "10" → "10"

  canonical_section = normalizeSection(section)
    # "1" → "1", "01" → "1", "02" → "2"

  canonical_slug = generateSlug(title)
    # "Core Concepts" → "core-concepts"
    # "What is React?" → "what-is-react"

  canonical_path = ".research/sections/{canonical_chapter}-{canonical_section}-{canonical_slug}/"

  # Tier 1: Exact canonical match
  tier1 = Glob("{canonical_path}research.md")
  if tier1 not empty:
    return { path: canonical_path, existing: true, tier: 1 }

  # Tier 2: Canonical chapter-section, any slug
  tier2 = Glob(".research/sections/{canonical_chapter}-{canonical_section}-*/research.md")
  if tier2 not empty:
    return { path: parent(tier2[0]), existing: true, tier: 2 }

  # Tier 3: Non-padded chapter variation
  raw_chapter = String(parseInt(chapter, 10))  # "01" → "1"
  tier3 = Glob(".research/sections/{raw_chapter}-{canonical_section}-*/research.md")
  if tier3 not empty:
    return { path: parent(tier3[0]), existing: true, tier: 3 }

  # Tier 4: Flexible pattern with first keyword
  keyword = canonical_slug.split('-')[0]
  tier4 = Glob(".research/sections/*-{canonical_section}-*{keyword}*/research.md")
  if tier4 not empty:
    return { path: parent(tier4[0]), existing: true, tier: 4 }

  # No match - use canonical for new directory
  return { path: canonical_path, existing: false, tier: "new" }