Signal Dreaming
v1.0.0Signal-driven memory consolidation for OpenClaw agents. Automatically consolidates recent session logs into long-term memory using recall frequency signals f...
⭐ 0· 45·0 current·0 all-time
bykelven.ling@lzyling
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the runtime instructions: the skill reads memory/.dreams/short-term-recall.json, daily logs, L2 topic files, and updates MEMORY.md and dream-log.md as described. It does not request unrelated binaries, environment variables, or credentials.
Instruction Scope
Instructions are focused on the memory/ area and the three-phase Sense/Consolidate/Settle workflow. Two items to be aware of: (1) automated cron runs use a delivery mode ('announce') which will publish a one-line dream summary to a configured channel — verify that channel before enabling automation to avoid accidental disclosure; (2) the SKILL.md suggests checking common skillDirs (e.g., ~/.openclaw/skills/) to find references/dream-protocol.md, so the agent may read files under the user's skills directory/home to locate the protocol file.
Install Mechanism
Instruction-only skill with no install spec and no code files — minimal disk/exec footprint and no external downloads. This is the lowest-risk install model.
Credentials
No environment variables or credentials are requested. All required inputs are local workspace files which are consistent with the stated purpose.
Persistence & Privilege
always is false and the skill does not request persistent platform privileges. It writes only to workspace memory files (L2 topic files, MEMORY.md, dream-log.md, .dream-backup.md) which is expected for a memory consolidation tool.
Assessment
This skill appears internally consistent and implements a local, file-based workflow. Before running it: (1) confirm the workspace path you give it is the one you expect (it will read and write files under memory/); (2) review or create MEMORY.md and memory/dream-log.md as recommended and back up any sensitive notes (the skill will rewrite MEMORY.md and append to dream-log.md); (3) if you enable automated runs, choose the delivery channel carefully — cron/announce will publish a one-line summary externally and could leak sensitive context; (4) be aware the skill will look for its protocol file in common skillDirs (including ~/.openclaw/skills), so check those paths if you have sensitive files there. If you need additional assurance, request a dry-run mode that outputs planned changes without writing files.Like a lobster shell, security has layers — review code before you run it.
automationvk97az20vjeqehct5g26p22fa7h84s57wconsolidationvk97az20vjeqehct5g26p22fa7h84s57wdreamingvk97az20vjeqehct5g26p22fa7h84s57wlatestvk97az20vjeqehct5g26p22fa7h84s57wmaintenancevk97az20vjeqehct5g26p22fa7h84s57wmemoryvk97az20vjeqehct5g26p22fa7h84s57w
License
MIT-0
Free to use, modify, and redistribute. No attribution required.