ClawHub

Persona Crafter

v1.0.0

Designing deeply nuanced AI identities with specific behavioral traits, values, and language patterns. Transform basic chatbots into authentic, opinionated c...

⭐ 0· 476·1 current·1 all-time
by@balkanblbn
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name, description, and SKILL.md all describe the same goal: guiding persona construction (diction, boundaries, values). The skill requires no binaries, env vars, or config paths — appropriate for a purely instructional persona guide.
ℹ
Instruction Scope
SKILL.md is concise and stays within persona design guidance; it does not instruct reading files, accessing credentials, or exfiltrating data. However the instructions are high-level and open-ended (e.g., "Give your agent a soul"), which grants the agent broad discretion when used — this is expected for a design guide but means downstream behavior depends on how the agent implements the guidance.
✓
Install Mechanism
There is no install spec or code to write to disk; the only install hint is a user-facing 'clawhub install' example in the doc. Instruction-only skills are lowest-risk from an install perspective.
✓
Credentials
The skill declares no required env vars, credentials, or config paths. No disproportionate access is requested.
✓
Persistence & Privilege
Flags show default behavior (always: false, agent invocation allowed). The skill does not request permanent presence or modify other skills; no elevated privileges are requested.
Assessment
This is a short, instruction-only persona-design guide and is coherent with what it claims. Technical risk is low because it asks for no credentials and installs no code. However, the guidance is deliberately high-level and can be used to create strong, opinionated agent personalities — which may lead to unexpected or undesired outputs (biased, controversial, or unsafe behavior) if not constrained. Before installing or enabling autonomous use: test personas in a safe sandbox, add explicit safety/refusal templates and content policy constraints, and limit autonomous invocation if you want human review of persona changes.

Like a lobster shell, security has layers — review code before you run it.

latestvk97cnbmvvgj3bbtjmgh3tw71xh823rpy

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments