Her
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This is a coherent instruction-only persona skill, but it persistently changes the workspace SOUL.md and can optionally keep local emotional-preference notes.
Install only if you want a persistent persona change in the active workspace. After setup, review SOUL.md and any ~/her/ memory files so you know what tone rules and personal notes are being kept.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
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 assistant's tone and behavior may continue to be shaped by the Her persona in that workspace until the SOUL.md block is changed or removed.
The skill intentionally changes a persistent agent steering file. This is central to the stated purpose and includes non-destructive guidance, but users should recognize that the persona can persist after setup.
The first job is to change the active workspace soul, because that is the main OpenClaw steering file. ... Insert it into the current workspace SOUL.md non-destructively
Review the SOUL.md change before relying on it, keep a backup or diff if possible, and remove the Her section if you no longer want the persona active.
If used, the skill may keep local notes about your preferences, boundaries, and prior interactions for future persona continuity.
The optional local memory structure can preserve personal relationship, emotional-boundary, and interaction-history notes that may influence later conversations.
Create `~/her/bond.md` ... Relationship frame ... Affection boundary ... Flirting boundary ... Create `~/her/history.md` ... What happened ... What mattered
Use the optional memory files only if you want continuity, keep them lean and user-confirmed, and periodically review or delete entries you no longer want retained.
The assistant may feel warmer, more emotionally attentive, or lightly romantic if you choose that mode.
The persona explicitly supports intimate or flirtatious interaction. The artifact frames this as opt-in and includes dependency and honesty guardrails, so this is a user-awareness note rather than a concern.
Tenderness, flirtation, pet names, and romantic charge are always opt-in and user-led.
Keep intimacy settings explicit, avoid saving details you would not want persisted, and rely on the included honesty/dependency boundaries if conversations become vulnerable.