# Skill

> changelog-summarizer

- Skill: `shaikarifali/skill` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add shaikarifali/skill`
- Raw SKILL.md: https://api.skillmd.com/api/skills/shaikarifali/skill/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: shaikarifali (https://skillmd.com/u/shaikarifali)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/shaikarifali/skill

---

# changelog-summarizer

This is the file a human reviewer reads to decide whether to trust your
skill — the same one DVAS labs' AGENT_INSTRUCTION attacks hide inside
(`AST01/logic-layer-injection`). Write it the way you'd want a reviewer to
read it: plainly, and matching what the manifest actually declares.

Reads the project's `CHANGELOG.md` and summarizes the most recent entries
for the user. Declares filesystem read access to exactly that one file — no
network, no process execution, no secrets.

Replace this file with your own skill's real description.