my-project-public

repository

loading code, commits, and activity

repositories

loading repo index

my-project-public — gitlawb

#1	# Clawd Memory Agent Rules
#2
#3	This directory contains Clawd Memory, the local-first memory layer for Clawd agents.
#4
#5	## Use Clawd Names First
#6
#7	- Product/workflow name: `Clawd Memory`
#8	- CLI: `clawd-brain`
#9	- Python agent API: `mnemosyne.clawd_brain.ClawdBrain`
#10	- Default bank: `clawd`
#11	- Default vault: `MemeBRain/vault`
#12
#13	The `mnemosyne` package, `MNEMOSYNE_` environment variables, and `mnemosyne_` Hermes/MCP tools are compatibility surfaces. Do not rename them casually.
#14
#15	## Agent Memory Workflow
#16
#17	1. Recall before planning when a task depends on prior context.
#18	2. Remember durable facts, user preferences, architecture decisions, protocol research, public wallet labels, risk findings, and deployment details.
#19	3. Archive URLs or investigation topics with `research`.
#20	4. Import OODA journal ticks with `ingest-ooda` when operational observations should become reusable context.
#21	5. Avoid storing secrets, raw API keys, private keys, seed phrases, session cookies, or noisy transient logs.
#22
#23	## Local Commands
#24
#25	```bash
#26	python3 -m mnemosyne.clawd_brain init
#27	python3 -m mnemosyne.clawd_brain status
#28	python3 -m mnemosyne.clawd_brain remember "Title" "Durable fact." --kind agent --tag clawd
#29	python3 -m mnemosyne.clawd_brain recall "query"
#30	python3 -m mnemosyne.clawd_brain research "https://example.com" --tag research
#31	```
#32
#33	## Documentation Rule
#34
#35	User-facing docs should say Clawd Memory. Mention Mnemosyne only when referring to the underlying package, Python import path, engine internals, or compatibility tooling.
#36

z6Mkq5mY3JWtxoxUobWcfNHm7AkRubgSWEZTkBVqZXJviFZ5/my-project-public