Retrospective
ReviewAudited by ClawScan on May 10, 2026.
Overview
This is a coherent retrospective skill, but it will read agent memory/config context and set up a recurring weekly retrospective job.
Install this only if you want a weekly retrospective that can read agent memory, cron/session/config context, write reports under memory/, and update MEMORY.md. Review the first generated report and the cron entry, avoid storing secrets in memory, and treat any skill-install or config recommendations as suggestions to approve manually.
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.
After first use, the retrospective may keep running weekly until the cron job is removed.
The skill creates a recurring autonomous agent turn. This is clearly disclosed, limited to a weekly retrospective, and uses an isolated session, so it is a persistence note rather than a concern.
Set up a weekly cron job on first use ... cron(action="add", job={ "name": "Weekly Retrospective" ... })Only use it if you want the recurring schedule; check the cron list after first use and remove or edit the job if it is not desired.
Generated retrospectives and MEMORY.md updates may include private work history, mistakes, blocked tasks, or configuration details that future agent sessions may reuse.
The skill reads private/persistent agent memory and writes new persistent memory. This is expected for a retrospective, but it can expose sensitive logs and shape future agent behavior.
Read `memory/` files ... Read `MEMORY.md` ... Write the retro to `memory/retro-YYYY-MM-DD.md` ... Update `MEMORY.md` with key findings
Review the generated retrospective and MEMORY.md changes, avoid storing secrets, and correct any inaccurate conclusions before relying on them later.
The report could mention account, session, API key, or integration problems, which may reveal operational details if shared.
The retrospective may inspect authentication/configuration status. The artifacts do not show credential collection, secret output, or transmission, but the area is sensitive.
Check session status for usage patterns ... Auth/credentials: Any expired, rotated, or missing? ... Tool config: Any env vars, API keys, or integrations that need attention?
Ensure outputs discuss credential status only, not secret values, and do not share generated retrospectives publicly without review.