Retrospective Agent
PassAudited by VirusTotal on May 11, 2026.
Overview
Type: OpenClaw Skill Name: retrospective-agent Version: 1.0.0 The retrospective-agent skill is a structured framework for logging execution lessons and process improvements. It contains strong safety guardrails in SKILL.md and references/boundaries.md that explicitly prohibit storing sensitive data (credentials, SSH keys), modifying core identity files, or performing unauthorized network/config changes. The skill operates entirely through transparent file logging within the workspace and shows no signs of malicious intent or high-risk vulnerabilities.
Findings (0)
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.
Future retrospectives or workflow recommendations may be shaped by prior logged lessons, including mistakes if they are logged too broadly or without enough context.
The skill stores reusable corrections and lessons in persistent workspace files, so those entries could later influence the agent's behavior or recommendations.
Operational data lives in: - `workspace/ops/retrospective-agent/` Expected first-pass files: - `workspace/ops/retrospective-agent/corrections.md` - `workspace/ops/retrospective-agent/weekly/` - `workspace/ops/retrospective-agent/domains/` - `workspace/ops/retrospective-agent/projects/`
Review logged lessons periodically, keep them scoped to the right project or domain, and avoid storing secrets, sensitive personal details, or one-off preferences as durable rules.