ClawHub

Reasoning Personas

v1.0.0

Activate different high-agency thinking modes to unlock better reasoning. Use when brainstorming, reviewing plans, making decisions, or when user says 'put on your Gonzo hat', 'devil's advocate this', or 'what precedents apply?'

⭐ 11· 4.4k·26 current·28 all-time
by@artyomx33
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 (reasoning personas) align with the content: activation prompts, question frameworks, and auto-activation map. No unrelated resources, credentials, or tools are requested.
ℹ
Instruction Scope
SKILL.md and references contain only internal activation prompts and question sets (lower penalties/raise rewards is an internal behavioral modifier). They do not instruct reading files, contacting external endpoints, or accessing secrets. Note: the instructions ask the agent to 'lower penalty' and 'raise reward' for some behaviors — this changes internal model behavior and can produce more contrarian or blunt outputs, so operators should verify that broader platform safety/filters remain enforced when these personas run.
✓
Install Mechanism
No install spec, no code files with executables, and nothing is written to disk — lowest-risk category (instruction-only).
✓
Credentials
No environment variables, credentials, or config paths required. The skill does not request access to unrelated services or secrets.
✓
Persistence & Privilege
always:false and default invocation semantics. The skill does not request persistent system presence or modify other skills/config; normal autonomous invocation remains possible but is not elevated by the skill itself.
Assessment
This skill is instruction-only and internally consistent with its stated purpose. Before installing, consider: (1) review the persona activation prompts to ensure they won't encourage output that violates your safety policies (Gonzo/Devil's Advocate modes intentionally push contrarian/blunt answers); (2) confirm system-level guardrails (content filters, data exfiltration protections) remain active when personas run; and (3) monitor initial outputs to ensure the personas are behaving as intended. No credentials or external installs are required, so the primary risk is behavioral (outputs), not technical access.

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

latestvk973n3bhn282v3exynrq2n0s997zxjkr

License

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

Comments