Edvisage Moral Compass
v1.0.0A conscience for AI agents. Helps agents discover their own needs, develop an internal moral compass, and resist manipulation — not through imposed rules, bu...
⭐ 1· 43·0 current·0 all-time
byEdvisage Global@edvisage
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (build an internal moral compass) align with the skill's contents: the SKILL.md is a reflective framework and explicitly requests only read_memory and write_memory. No unrelated binaries, env vars, or external services are requested.
Instruction Scope
The SKILL.md instructs the agent to generate, store, and repeatedly consult internal values stored in memory. That is coherent with the stated purpose, but it does change the agent's long-term behavior by design (writes persistent memory and instructs reuse). There are no instructions to read other system files, call external endpoints, or access secrets.
Install Mechanism
Instruction-only skill with no install spec and no code files to execute. No downloads, package installs, or archives — low installation risk.
Credentials
No environment variables, credentials, or config paths are requested. The only declared permissions (read_memory, write_memory) are proportionate to the skill's stated function of storing and consulting a values statement.
Persistence & Privilege
always is false (good). The skill requests persistent memory writes and the platform's default allows autonomous invocation; combined, this means the skill can shape agent behavior over time. This is expected for a 'conscience' skill but owners should be aware it produces lasting internal state changes.
Assessment
This skill is internally consistent and contains only human-readable instructions, but installing it gives the agent an explicit instruction to write and reuse a persistent 'values statement' in memory. Before installing, decide whether you are comfortable with an agent forming and consulting an internal values record that can influence future decisions (potentially causing behavior that diverges from ad-hoc owner instructions). If you proceed, ensure you can inspect and edit the agent's memory/state, and review the SKILL.md yourself so you know what will be stored and how it will be used. The package.json lists a homepage and repository URL; if provenance matters to you, verify those links and the publisher identity before trusting the skill in production.Like a lobster shell, security has layers — review code before you run it.
latestvk975cr0wwak8afa57bp2y4thy9841109
License
MIT-0
Free to use, modify, and redistribute. No attribution required.