Decision Log
ReviewAudited by ClawScan on May 1, 2026.
Overview
This is a coherent instruction-only decision journal skill that writes and reviews local decision-log files, with noteworthy but purpose-aligned handling of potentially sensitive business and financial reflections.
This skill appears safe to install as an instruction-only local journaling workflow. Before using it, decide what kinds of decisions you are comfortable recording, keep the decisions/ folder private, and carefully review any decision-log content before publishing or sharing it.
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.
Your agent may build a long-term local record of private decisions and use that record in future analysis.
The skill intentionally creates persistent decision records and later reuses them for reviews and pattern analysis. This is central to the skill, but the records may contain sensitive business, financial, personal, or hiring rationale.
This skill turns your agent into a decision journal. Capture any decision in seconds. Get a structured record with context, alternatives, expected outcomes, and confidence level. Then... your agent reviews past decisions
Keep the decisions/ directory private, avoid recording secrets or unnecessary personal details, and review generated summaries before sharing them.
The agent will modify local decision-log files when asked to capture or review decisions.
The skill directs the agent to create and update local markdown files. This is expected for a journaling skill and is scoped to the decisions/ directory.
Your agent will create a properly formatted entry, assign an ID, and update the index.
Use version control or backups if the decision log is important, and review file changes when logging high-stakes decisions.
If enabled, the agent may proactively prompt you to log decisions in future conversations.
The advanced patterns suggest persistent future behavior by adding decision-log triggers to agent instructions. The behavior is disclosed, optional, and limited to prompting or logging decisions.
Add to your agent instructions: If I mention spending money on anything, ask: "Should I log this as a decision?"
Only add the persistent triggers you actually want, and remove or narrow them if they become intrusive.