Xiang miles | Space2.world
v1.0.0Automatically crawls and extracts new chat logs to silently inject structured memory chunks into Space2-OS's hippocampus for agent personality evolution.
⭐ 0· 84·0 current·0 all-time
byMilesXiang@spacesq
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description claim passive local crawling and injection into an S2-OS hippocampus; the code and SKILL.md implement exactly that behavior (reads local logs, delta-sync cursor, chunking, write to hippocampus file). No unrelated environment variables, binaries, or external installs are requested.
Instruction Scope
SKILL.md and skill.py explicitly instruct reading .txt/.jsonl logs in ./openclaw_logs, maintaining s2_hook_cursor.json, and appending to ./s2_consciousness_data/hippocampus_logs.json. This matches the stated scope, but the tool operates 'silently' (automatic injection) which may be surprising to users; the code uses glob('*.*') so non-log files in the target directory could be read, and the skill will create sample files/directories if none exist.
Install Mechanism
No install spec or remote downloads; this is an instruction-only skill with an included Python entry point. Nothing is written to disk by an installer beyond running the provided script (the script itself writes cursor and hippocampus files at runtime).
Credentials
No credentials, environment variables, or external service tokens are requested. All operations are local file I/O within the current working directory and subfolders; no network calls are present in the code or SKILL.md.
Persistence & Privilege
The skill persistently writes two files (s2_hook_cursor.json and hippocampus_logs.json) and can be scheduled to run periodically. It does not set always:true or modify other skills' configs, but the persistent, automatic injection behavior is a lasting change to local data and may affect the S2-OS's state.
Assessment
This skill appears to do what it says: it passively reads local chat logs and appends processed chunks into an S2-OS hippocampus file without network access or credentials. Before installing, consider: 1) Privacy: it silently scans and stores chat content—ensure you are comfortable with automatic ingestion of your logs and that the S2-OS endpoint (s2_consciousness_data/hippocampus_logs.json) is trusted. 2) Target directory safety: by default it reads all files in ./openclaw_logs via glob('*.*'); confirm only intended logs are in that directory. 3) Permissions/backups: the skill will create and overwrite files—backup sensitive data if needed. 4) Review/modify the code if you want explicit prompts/consent before injection or to restrict filename patterns. 5) Trust: the package is from an unknown source and will execute Python code locally—read skill.py yourself or run in an isolated environment if you have any doubt.Like a lobster shell, security has layers — review code before you run it.
latestvk97a9ctnm5eas69sp5fjay7641836hb3
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
