From memory-bank
List memories stored in the GCP Reasoning Engine Memory Bank. Use when the user wants to see what's in their memory bank, filter by scope, or look up a memory ID before updating or deleting.
How this skill is triggered — by the user, by Claude, or both
Slash command
/memory-bank:memories-listThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
List memories from the GCP Memory Bank.
List memories from the GCP Memory Bank.
Locate the script — it lives in a different place per runtime:
ls -d ~/.claude/scripts/memory-bank/list_memories.py \
~/.gemini/config/plugins/memory-bank/scripts/list_memories.py \
~/.claude/plugins/cache/*/memory-bank/*/scripts/list_memories.py 2>/dev/null
Run the first match in this precedence order, which is not the order ls
prints (it sorts):
~/.claude/scripts/memory-bank/… — the version-free symlink Claude Code's
install-symlinks.sh maintains.~/.gemini/config/plugins/memory-bank/scripts/… — where Antigravity
installs the plugin.~/.claude/plugins/cache/ — last resort if the symlink was
never created; prefer the highest version number.python3 <the path you chose> [--scope current|all]
Default: --scope current (shows global + project memories for the current workspace).
If step 1 printed nothing, stop and report that the memory-bank plugin does
not appear to be installed. Do not use $CLAUDE_PLUGIN_ROOT — it is set for
hooks only and is empty in a command you run — and do not collapse this into
one line with $(…), which gets auto-denied in a headless Antigravity session.
Format the output as a Markdown table: | # | ID | Scope | Fact |
--scope all only if the user explicitly asks to see all memories regardless of project.npx claudepluginhub mlarkin00/plugins --plugin memory-bankGuides completion of development work by verifying tests, detecting environment, and presenting structured options for merge, PR, or cleanup.
Guides creation and editing of skills using test-driven development with pressure scenarios and subagents to verify agent compliance.
Dispatches multiple subagents concurrently for independent tasks without shared state. Use when facing 2+ unrelated failures or subsystems that can be investigated in parallel.