Weekly Retro

Weekly retrospective that analyzes memory logs to identify accomplishments, recurring patterns, friction points, and forward-looking recommendations. More st...

This skill appears to do what it says: read local memory logs, analyze patterns, and produce a markdown retrospective. Before running it: 1) Inspect the memory files that will be read (default ~/.openclaw/workspace/memory) for any secrets or sensitive content you don't want included; 2) Run the pipeline against a small test directory first (use --memory-dir to point to a safe test folder) to see output format; 3) Note that analyze.py can read SOUL.md and AGENTS.md if you pass those paths or configure them — don't point it at files with credentials; 4) The skill writes reports and a history file into your workspace (vault/history paths are configurable) — ensure those output locations are acceptable; 5) There is no network/exfiltration observed in the code, but always run third-party scripts in a controlled environment if you have high sensitivity requirements.

Type: OpenClaw Skill Name: weekly-retro Version: 1.0.0 The weekly-retro skill bundle is a well-documented tool designed to analyze local memory logs and generate strategic retrospectives within the OpenClaw workspace. The Python scripts (gather_week.py, analyze.py, retrospective.py, and history.py) operate entirely on local data using the standard library, with no evidence of network activity, data exfiltration, or malicious execution. The instructions in SKILL.md are clearly aligned with the tool's purpose of summarizing accomplishments and friction points from the user's workspace files (~/.openclaw/workspace/memory).