People Memories
ReviewAudited by ClawScan on May 10, 2026.
Overview
The core people-memory feature is understandable, but it needs review because it quietly persists personal details and describes an automatic Telegram reminder job without clear setup or control boundaries.
Install this only if you intentionally want a persistent local memory about people. After installation, check ~/.clawdbot/people-memory.json and any logs for saved notes, confirm whether a cron or Telegram notification job exists, and avoid storing sensitive third-party information unless you have clear deletion and retention controls.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Names, preferences, birthdays, anniversaries, and other personal context may be saved and reused later without obvious per-note confirmation.
The skill stores personal details in a persistent memory file and can populate it automatically from voice/chat cues with low visibility.
`~/.clawdbot/people-memory.json` now stores ... `notes` ... When you say something like “remember Alex likes cats,” it automatically runs the `remember` command ... confirmations quiet
Use only if you want this kind of persistent people memory. Ask for explicit save confirmations, review the stored JSON file, and require clear retention and deletion controls.
A background job could continue sending reminder information about people outside the immediate chat session.
This describes autonomous daily background behavior and external notification delivery, but the artifacts do not show a clear install spec, schedule opt-in, Telegram credential declaration, or disable path.
A helper cron job runs `python3 skills/people-memories/scripts/people_memory.py reminders --days 0 --window 7 --format message` each morning and delivers the resulting digest over Telegram
Before installing, confirm whether any cron job or Telegram integration is actually created, how to disable it, and what data it sends.
The skill will run local Python code when matching voice transcript text is detected.
The extension executes the bundled Python script to save a memory. This is expected for the skill’s design and uses execFile with structured arguments, but it still means installing the skill enables local code execution.
execFile("python3", args, { stdio: "ignore" }, (err) => {Review the bundled script before enabling the extension and install only from a source you trust.