chronicle-write

name	chronicle-write
description	Guided ritual for writing to project memory files. Use before any write to PROJECT_MEMORY.md, UPDATE_LOG.md, or PHASE_STATUS.md to ensure correct timestamp, format, and entry ordering.

chronicle:write — Memory Write Ritual

Step 1 — Get a fresh timestamp

Never hand-type. Never reuse the session-start timestamp.

# Any project (chronicle portable):
python3 ~/.claude/plugins/chronicle/scripts/timestamp.py

# DocinaBox (project-local, preferred when present):
python3 scripts/agent_harness/current_timestamp.py

Copy the output verbatim. Example output: Tuesday, 24-06-2026, 3:15 pm

Step 2 — Update the file header

Replace the Last Updated: line at the very top of the target file:

Last Updated: Tuesday, 24-06-2026, 3:15 pm

Step 3 — Build your entry

Two layers, no subheading between them:

## Tuesday, 24-06-2026, 3:15 pm, [branch-name] Short title

- Plain-language bullet (outcome, decision, or risk — readable by a stakeholder)
- Plain-language bullet 2 (if needed, max 4 total)

- Technical detail: file:line, migration name, API contract, failure mode
- Technical detail 2 (if needed, max 4 total)

Format rules:

• Day is spelled out: Tuesday not Tue
• Date is DD-MM-YYYY: 24-06-2026
• Time has no leading zero, lowercase am/pm: 3:15 pm not 03:15 PM
• Branch name in square brackets: [dev] not <dev>
• Title describes WHAT and WHY, not the process ("Added X to fix Y" not "Updated file")
• No process narration: no "spawned agents", "wave 3", model choices, tool counts

Step 4 — Prepend the entry

Insert your new ## entry immediately after the Last Updated: line (and any --- separator). Newest entries are always at the top of the file.

Step 5 — Sync (recommended)

After writing, propagate to both native stores so the other agent sees it next session:

python3 ~/.claude/plugins/chronicle/scripts/sync.py

Or invoke chronicle:sync if available.