Continuity Framework
AdvisoryAudited by Static analysis on Apr 30, 2026.
Overview
No suspicious patterns detected.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may retain and later surface personal or inferred context, including relationship dynamics or significant moments.
The skill explicitly stores long-term or permanent memories, including relationship and episodic information, which may be reused in later sessions.
| `relationship` | Connection dynamics | Long-term | ... | `moment` | Significant episodes | Permanent |
Review the generated memory files periodically, and avoid using the skill for conversations whose details should not become persistent context.
Local files may accumulate persistent agent memory outside the immediate chat session.
The script persists questions, identity, and reflection logs to a local memory directory by default.
MEMORY_DIR = Path(os.environ.get("CONTINUITY_MEMORY_DIR", Path.home() / "clawd" / "memory"))Know the configured memory directory, back it up or delete it according to your privacy needs, and set CONTINUITY_MEMORY_DIR if you want a different storage location.
If the heartbeat integration is added, reflection may run automatically after conversations and update persistent memory.
The skill recommends a heartbeat integration that would run reflection after idle sessions rather than only when manually invoked.
**Trigger**: Heartbeat after conversation idle > 30 minutes **Action**: Run continuity reflect
Only add the heartbeat trigger if you want automatic reflection, and keep the idle threshold and generated memory output under your control.