Osop

OSOP workflow authoring, validation, risk analysis, and self-optimization for AI agents

This skill appears coherent with its stated purpose, but take these precautions before installing: (1) Inspect ~/.osop/config.yaml for any stored secrets or tokens and remove or rotate them if you don't trust the publisher. (2) Verify the OSOP_MCP_URL endpoint (is it an official/owned service?) — avoid pointing it to untrusted servers that could receive logs or workflow contents. (3) Remember that executing workflows can run 'cli', 'db', 'docker', 'infra' steps defined in the .osop file; only run workflows you’ve reviewed or that include approval gates. (4) Installing pyyaml is low-risk but will add a Python dependency; if you have strict supply-chain requirements, consider installing it in an isolated environment. If you need higher assurance, ask the publisher for source code or an official package release to audit how the MCP endpoint and config are used.

Type: OpenClaw Skill Name: osop Version: 1.2.0 The OSOP skill bundle is a framework for defining and managing AI agent workflows (Standard Operating Procedures). The SKILL.md file provides instructions for authoring, validating, and risk-assessing workflows, including specific sub-skills for security reviews (/osop-review) and execution logging. The bundle lacks executable code and its instructions are strictly aligned with its stated purpose of workflow orchestration, incorporating safety features like dry-run modes and approval gates.