Dispatch a Haiku sub-agent to search hmem and return up to 10 memory summaries.
npx skills add https://github.com/YuryZZZ/opencode-global --skill hmem-recall
Please help me install this Agent Skill. Skill: hmem-recall Source: https://github.com/YuryZZZ/opencode-global/tree/main/skills/hmem-recall Command: npx skills add https://github.com/YuryZZZ/opencode-global --skill hmem-recall
Dispatch a Haiku sub-agent to search hmem for relevant memories. Sub-agent returns matching entries as ID + one-line summary. Main agent context stays clean.
Define QUERY and TYPE, then dispatch the Haiku sub-agent to search hmem and return up to 10 matching entries.
High-intent search queries and answers about installing and using this skill.
Search past session context by dispatching a Haiku sub-agent to query memory, returning up to 10 deduplicated entries as compact ID and summary pairs to keep the main agent context clean.
Retrieve past lessons without an exact memory ID by dispatching a Haiku sub-agent to perform keyword and semantic search, formatting results into a one-line summary for quick reading.
Yes, you can use structured prompts to search hmem by dispatching a Haiku sub-agent, which deduplicates results by ID and limits output to up to 10 matching entries.
Yes, memory recall with a Haiku sub-agent limits search results to up to 10 entries, deduplicating by ID and returning them in a compact summary format for easy ingestion.
sigridjinethPersist user preferences and past interactions across AI sessions with structured memory.
yairbedermanLog decisions and project lessons to preserve session context across work tasks.
HerbertJulioCapture user decisions, preferences, and lessons into a memory index.
ironkid90Prune and prioritize DeepSeek context messages using semantic deduplication and relevance scoring.
