memory · sametcelikbicak — Claude Code command: what it does, how to install
memory is a Claude Code command for reading, creating, or compacting project memory. It works with files under .team/memory/.
MIT ★ 71 updated 2026-11-13
.claude/commands/opencode-team-memory.mdWhat memory does
The memory command summarizes project memory, creates missing files, or reduces the size of its index and logs. With no argument, it reads INDEX.md, reports log sizes, and lists recent decisions and open questions.
Use init to create the memory tree, missing logs, and .team/.gitignore containing * if that file is absent. Existing content stays untouched. The compact argument archives excess log lines and trims INDEX.md. It refuses to run while subagents are active; if more than 60 unique index lines remain, it stops and reports the count.
What it helps with
- inspect the project's memory summary
- create missing memory files
- compact memory logs and the index
Example prompts for the agent
- Show me the current project memory status.
- Create the memory tree and any missing logs.
- Archive excess log lines and compact the index.
How to install memory
Manually: put the files in place
.claude/commands/opencode-team-memory.mdThe built-in /memory overrides a file with the same name, so the file is renamed. Call it with /opencode-team-memory. For all projects use ~/.claude/commands/
Manually: put the files in place
~/.codex/prompts/memory.mdCall it with /prompts:memory. Codex marks this format deprecated in favour of skills
Manually: put the files in place
.github/prompts/memory.prompt.mdManually: put the files in place
.gemini/commands/memory.tomlGemini stores commands as TOML: the prompt text goes into the prompt field
Questions about memory
How do I install memory in Claude Code?
Add the command file at .claude/commands/opencode-team-memory.md.
What does the memory command do?
It inspects project memory under .team/memory/, creates missing files, or compacts the index and logs.
When will memory refuse to compact?
The compact action stops while subagents are active. It also stops and reports the count if more than 60 unique lines remain in the index.
Source text
memory.md · MIT
--- name: memory description: Inspect, seed, or compact project-local agent memory. agent: orchestrator --- Inspect, seed, or compact project-local memory in `.team/memory/`. Arguments: - `status` (default): read-only summary of `INDEX.md`, log sizes, latest decisions, and open questions. - `init`: create the memory tree and missing logs, plus `.team/.gitignore` (`*`) when absent; never overwrite existing content. - `compact`: archive overflow log lines and shrink `INDEX.md`; refuse when subagents are active. If unique index lines still exceed 60, stop and report the count instead of dropping facts. $ARGUMENTS
© sametcelikbicak · MIT · sametcelikbicak/opencode-team
Similar slash commands
All slash commands for Agent setup: memory, context, orchestration →memory by other authors comes first.