| name | log-writer |
| description | Centralized writer for `<config-root>/memory/log.md` — the unified append-only chronicle of audit-worthy Nucleus operations (v4.7.1+, centralized in v4.7.2).
Auto-fires when ANOTHER cortex command finishes its main work and needs to append a log entry. Not invoked by the user directly; invoked by /listen, /morning, /end-day, /end-week, /reindex, /research-gaps, /merge-research-draft, /cleanup, /rehearse at their respective "Log to chronicle" steps.
Inputs: op_name (the command name, e.g., "listen"), summary (the one-line description for the entry). Optionally body (1-2 line context block).
Output: appends one entry to `<config-root>/memory/log.md`. Creates the file with the standard header if it doesn't exist. Never modifies prior entries.
Format is defined here in one place. If the format changes, this skill updates; callers don't need to change their instruction strings.
|
log-writer
Centralized append-only writer for the Nucleus operations chronicle.
When fired
By another cortex command at the end of its main workflow, when that command needs to log its completion. Callers include:
/listen Step 7.5/morning Step 4.5/end-day Step 5.7/end-week Step 5.7/reindex Step 5.5/research-gaps Step 4.5/merge-research-draft Step 4.5/cleanup Step 4.7/rehearse Step 4.5- (future)
/forget and /remember --full may opt in
This skill is not auto-fired by user utterances or trigger phrases. It's a programmatic primitive called from other commands.
Inputs
The calling command provides:
op_name (required): the command name without the leading slash. E.g., listen, morning, end-day.summary (required): one-line description of what this run accomplished. Should be informative enough to be useful when grepped months later.body (optional): 1-2 line context block indented under the entry. Used rarely; only when the summary needs unpacking.timestamp (optional): override the timestamp. Default is current local time in the user's identity time zone.
Procedure
-
Resolve <config-root> via the standard pattern (~/Documents/.claude-plugin-config-root).
-
Determine timestamp. If timestamp was provided, use it. Otherwise read <config-root>/identity.md for time zone and use current local time. Format: YYYY-MM-DD HH:MM (minute precision; no seconds).
-
Ensure the log file exists. Path: <config-root>/memory/log.md. If missing, create with this header:
# Memory log
_Append-only chronicle of audit-worthy Nucleus operations. Cortex writes; humans read. See `references/log-chronicle.md` for the spec._
-
Compose the entry. Format:
## [<timestamp>] <op_name> | <summary>
If body was provided, append it on the next line, indented (4 spaces):
## [<timestamp>] <op_name> | <summary>
<body line 1>
<body line 2>
-
Append. Open the file in append mode, write the entry. Never read-and-rewrite — that would invite race conditions if two scheduled commands fire simultaneously.
-
Best-effort. If the write fails (disk full, permissions, etc.), the caller's main work must NOT be marked failed. Log-writer returns a status to the caller; the caller continues regardless.
Format consistency
All callers produce entries with the same structure:
## [YYYY-MM-DD HH:MM] <op_name> | <one-line summary>
If this format ever needs to change (e.g., add structured fields, add an op-type prefix), update this skill in one place. Callers' instructions just say "invoke log-writer with op_name=X, summary=Y" — they don't repeat the format string.
What this skill does NOT do
- Does not modify or delete prior log entries. Append-only.
- Does not enforce retention. That's the caller's job or the user's via
cortex.user-context.md log_chronicle.max_entries (future).
- Does not deduplicate. If two commands log the same op at the same minute, both entries are written.
- Does not interpret
summary or body — opaque strings to log-writer.
- Does not fire on user trigger phrases. Programmatic invocation only.
- Does not log itself. There's no "## log-writer fired" entry; the log is for audit-worthy user operations, not internal plumbing.
