Context Engineering (Koylan)
PassAudited by ClawScan on May 1, 2026.
Overview
This instruction-only skill is coherent and purpose-aligned, but it relies on mutable GitHub-hosted sub-skill documentation that users should trust before use.
This appears safe as an instruction-only routing skill, but it depends on sub-skill files fetched from a GitHub repository at runtime. Use it if you trust that upstream source, and consider reviewing or pinning the referenced sub-skill documentation for stable behavior.
Findings (2)
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.
The agent may use guidance that is not included in this package and that can change over time if the upstream repository changes.
The primary sub-skill content is fetched from a mutable GitHub main branch rather than being fully included or pinned in the reviewed artifact.
Sub-skill SKILL.md files are loaded directly from GitHub: https://raw.githubusercontent.com/muratcankoylan/Agent-Skills-for-Context-Engineering/main/skills/<sub-skill>/SKILL.md
Install only if you trust the referenced GitHub repository; for stronger reproducibility, prefer a pinned commit or locally reviewed copy of the sub-skill files.
If the external sub-skill text changes or contains unsafe guidance, it could steer the agent's response during context-engineering tasks.
The skill makes externally loaded sub-skill instructions part of the agent's operating guidance. This is purpose-aligned, but retrieved instructions should not override user intent or higher-priority safety rules.
Match the user's task to one sub-skill and read its SKILL.md before proceeding... Follow the instructions in that file.
Treat the fetched sub-skill files as advisory documentation and ensure they remain subordinate to the user's request and platform policies.