Capture and categorize agent actions, escalations and failures into dated markdown files.
npx skills add https://github.com/mentilead/growthOS --skill log-mentilead
Please help me install this Agent Skill. Skill: log Source: https://github.com/mentilead/growthOS/tree/main/skills/observation-logger Command: npx skills add https://github.com/mentilead/growthOS --skill log-mentilead
This Skill provides a structured way to log important observations, decisions, and outcomes during an experiment or project, ensuring that valuable insights are captured and can be reviewed later.
Use the log skill to quickly capture what just happened.
High-intent search queries and answers about installing and using this skill.
To track experiment observations and agent actions in markdown files, this Skill captures and categorizes events by reading and writing persistent state to STATUS.md and MEMORY.md. It ensures valuable insights and outcomes are recorded for later analysis.
The best way to log autonomous agent decisions during a project lifecycle is to capture the action, its outcome, and its significance. This Skill categorizes these observations and escalations to maintain a structured markdown record for analysis.
Yes, you can generate monthly summaries of logged events and autonomy progress. This Skill reviews captured observations and experiment tracking data to summarize outcomes and reflect on past project decisions.
To add deeper context and reasoning to past experiment tracking logs, use the reflection feature. It allows you to append additional insights to existing observations stored in your project markdown files.
Yes, you need specific markdown files to manage persistent state for experiment tracking. This Skill requires reading and writing to STATUS.md, MEMORY.md, and related markdown files to log observations and score agent autonomy.
kimon1230Save session progress and project state to persistent status files.
netanel-abergelEvaluate AI personal assistant performance with structured scoring and owner feedback.
EmasoftDetect and classify remote agent failures during heartbeat checks and message delivery.
mentileadSynthesize experiment observations and metrics into publication-ready content for Substack, LinkedIn, and Medium.
