Diagnose and configure MemSearch memory behavior. Use when the user asks about MemSearch configuration, plugin summarization, PROJECT.md/USER.md maintenance, memory directories, index health, provider routing, prompt files, or migration/compatibility questions.
81%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Critical
Do not install without reviewing
The canonical home for this skill is memory-config in zilliztech/memsearch
Detected a prompt injection in the skill instructions. The skill contains hidden or deceptive instructions that fall outside its stated purpose and attempt to override the agent’s safety guidelines or intended behavior.
The skill is a MemSearch configuration guide with no deception, obfuscation, or covert code, but its DSH section offers to enable saved credentials for all headless jobs, which widens credential exposure beyond what the summarization task strictly needs.
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The required diagnostic reads the user's own MemSearch daily memory files and index state, which are LLM summaries of the user's sessions and could carry text from web pages the agent fetched earlier, so an outsider could only reach this text by first getting the agent to fetch their page.
zilliztech/memsearch
3e296a3
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.