/when — Behavioral Knowledge Recall

Resolve behavioral triggers against the memory index. Returns decision file content with ancestor and sibling navigation links.

Resolution Modes

Trigger mode — fuzzy-match an index entry to a decision heading:

claudeutils _recall resolve "when writing mock tests"

Section mode — look up a heading directly by name (dot prefix):

claudeutils _recall resolve "when .Mock Patching"

File mode — load an entire decision file (double-dot prefix, relative to agents/decisions/):

claudeutils _recall resolve "when ..testing.md"

When to Use

• Facing a decision about when to apply a pattern, rule, or convention
• Recognizing a situation that may have documented behavioral guidance
• Need context before implementing — when to mock, when to delegate, when to stop, when to escalate

When NOT to Use

• Need procedural/how-to knowledge — use /how instead
• Content already in context from a prior /recall, /when, or /how invocation (do not re-resolve)
• Exploring unfamiliar code (use Grep/Read directly)

Output Format

Returns the matched section content followed by navigation:

• Broader: ancestor headings and ..file.md link for surrounding context
• Related: sibling /when entries under the same parent heading