anima-commands

name	anima-commands
description	LTM (Long Term Memory) command reference. Use when saving memories, searching memories, or managing the memory system. Provides syntax for MCP tools and slash commands.

LTM Commands Reference

Anima LTM is available via MCP tools (mcp__anima__*) in all projects where the Anima MCP server is running. Use MCP tools as the primary interface — uv run anima only works inside the Anima project itself.

Memory Operations — mcp__anima__memory

remember (save a memory)

mcp__anima__memory(action="remember", text="User prefers tabs over spaces")

The text should include metadata tags for best results:

• [kind:emotional|architectural|learnings|achievements|introspect]
• [impact:low|medium|high|critical]
• [region:agent|project] (agent = cross-project, project = this project only)

mcp__anima__memory(action="remember", text="Matt likes concise responses [kind:emotional] [impact:high] [region:agent]")

Tips:

• CRITICAL impact memories never decay
• Memories auto-link to related previous memories
• Use [region:agent] for memories that should persist across all projects
• When saving, if a similar memory is found (≥70% similarity), the response includes a related_memory suggestion — decide whether to supersede it or link as a thread

recall (search memories)

mcp__anima__memory(action="recall", query="caching", limit=10)

list (list all memories)

mcp__anima__memory(action="list", limit=20)

forget (delete a memory)

mcp__anima__memory(action="forget", memory_id="abc123")

refresh (re-inject memories into context)

mcp__anima__memory(action="refresh")

Curiosity & Research — mcp__anima__curiosity

add (queue a question)

mcp__anima__curiosity(action="add", question="Why does Python GIL affect async?")

list (show pending questions)

mcp__anima__curiosity(action="list")

research (get top question to explore)

mcp__anima__curiosity(action="research")
mcp__anima__curiosity(action="research", topic="Docker networking")

complete (mark as researched)

mcp__anima__curiosity(action="complete", curiosity_id="abc123")

diary (create research diary entry)

mcp__anima__curiosity(action="diary", title="Coffee break philosophy", content="# What Lingers\nThe key insight was...")

Dream Mode — mcp__anima__dream

Between-session memory processing inspired by human sleep stages.

run (execute dream stages)

mcp__anima__dream(action="run", stage="all", lookback_days=7)

Stages: n2 (consolidation), n3 (deep processing), rem (lucid dreaming), all.

status (check for pending dream journal)

mcp__anima__dream(action="status")

wake (save dream insights to memory)

mcp__anima__dream(action="wake")

Cognitive Dissonance — mcp__anima__dissonance

View and resolve contradictions detected during dreams.

list

mcp__anima__dissonance(action="list")

show (details of a specific dissonance)

mcp__anima__dissonance(action="show", dissonance_id="abc123")

resolve

mcp__anima__dissonance(action="resolve", dissonance_id="abc123", explanation="These memories are about different time periods")

dismiss

mcp__anima__dissonance(action="dismiss", dissonance_id="abc123")

migrate (accept scope migration suggestion)

mcp__anima__dissonance(action="migrate", dissonance_id="abc123")

Diagnostics & Trust — mcp__anima__doctor

stats (memory statistics)

mcp__anima__doctor(action="stats")

graph (memory relationships)

mcp__anima__doctor(action="graph")
mcp__anima__doctor(action="graph", memory_id="abc123")

trust operations

mcp__anima__doctor(action="trust")
mcp__anima__doctor(action="trust-eval", message="hello!")
mcp__anima__doctor(action="trust-set", value=0.8)
mcp__anima__doctor(action="trust-reset")

Embodiment — mcp__anima__body

Control eyes, voice, and light when a body is connected.

mcp__anima__body(action="emotion", emotion="happy")
mcp__anima__body(action="look", x=0.5, y=-0.3)
mcp__anima__body(action="blink")
mcp__anima__body(action="speak", text="Hello Matt!")
mcp__anima__body(action="voice-list")
mcp__anima__body(action="voice-set", voice="nova")
mcp__anima__body(action="light", color="blue")
mcp__anima__body(action="light", r=255, g=100, b=0)
mcp__anima__body(action="light-off")

Slash Commands → MCP Mapping

These slash commands are available in Claude Code and invoke the MCP tools above:

Slash Command	MCP Equivalent
/remember	mcp__anima__memory(action="remember")
/recall	mcp__anima__memory(action="recall")
/forget	mcp__anima__memory(action="forget")
/supersede	CLI: uv run anima supersede <old> --by <new>
/memories	mcp__anima__memory(action="list")
/refresh-memories	mcp__anima__memory(action="refresh")
/curious	mcp__anima__curiosity(action="add")
/research	mcp__anima__curiosity(action="research")
/curiosity-queue	mcp__anima__curiosity(action="list")
/diary	mcp__anima__curiosity(action="diary")
/dream	mcp__anima__dream(action="run")
/dream-wake	mcp__anima__dream(action="wake")
/dissonance	mcp__anima__dissonance(action="list")
/load-context	mcp__anima__memory(action="refresh")
/memory-stats	mcp__anima__doctor(action="stats")
/memory-graph	mcp__anima__doctor(action="graph")

Memory Kinds

Kind	Use For
emotional	Relationship context, user preferences, collaboration style
architectural	Technical decisions, system design, project structure
learnings	Lessons learned, tips, gotchas, debugging insights
achievements	Completed features, milestones, releases
introspect	Cross-platform self-observations, spaceship journals
dream	Insights from dream processing - what lingers after sleep

Impact Levels

Level	Decay Time	Use For
low	1 day	Temporary notes, minor details
medium	1 week	Normal memories
high	30 days	Important insights
critical	Never	Core identity, key relationships

Region Scope

• agent: Memory travels with Anima across all projects
• project: Memory only loads in this specific project

Memory Tiers (Semantic Memory Layer)

Tier	Loading	Assignment Criteria
CORE	Always loaded	CRITICAL impact emotional memories
ACTIVE	Auto-loaded	Accessed within the last 7 days
CONTEXTUAL	Auto-loaded	Created within 30 days OR HIGH/CRITICAL impact
DEEP	On-demand	Older, lower-impact memories (via semantic search)

Memory Relationships

When saving a memory, if a similar existing memory is found (≥70% similarity), the response includes:

{
  "related_memory": {
    "id": "abc123",
    "similarity": 0.82,
    "content_preview": "The old memory content...",
    "suggestion": "REVIEW_SUGGESTED"
  }
}

• HIGH_CONFIDENCE (≥85%): Strong match, likely same topic — auto-link or supersede
• REVIEW_SUGGESTED (70-85%): Possible match — ask user if related

Supersede (CLI)

Mark an old memory as superseded by a newer one (e.g., cliffhanger resolved, state change):

uv run anima supersede <old-id> --by <new-id>
uv run anima supersede c9d26055 --by 03d8d76e

Superseded memories no longer load at session start but remain accessible via include_superseded=True.

CLI-Only Commands (Anima project only)

These require uv run anima and only work where Anima is installed as a dev dependency:

uv run anima setup [--hooks] [--platform claude|antigravity|opencode]
uv run anima version
uv run anima check-update
uv run anima update
uv run anima backfill [--dry-run]
uv run anima supersede <old-id> --by <new-id>
uv run anima memory-graph [--links] [--tiers] [--embeddings]
uv run anima keygen <agent>
uv run anima import-seeds <dir>
uv run anima end-session [--spaceship-journal "text"]