ClawHub
Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Chitin — Personality Persistence for AI Agents

v1.4.5

Personality persistence for AI agents. Remember how you think, not just what happened. Structured insights that survive session restarts.

⭐ 1· 2.7k·3 current·3 all-time
by@morpheis
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Benign
medium confidence
✓
Purpose & Capability
Name/description match the contents of SKILL.md: the guide describes installing and using a chitin CLI to store and retrieve structured 'insights' and triggers. There are no unrelated environment variables, binaries, or install steps declared in the registry metadata that would contradict the stated purpose.
ℹ
Instruction Scope
Instructions direct the agent/operator to install a CLI, initialize a local database, import/export insight files, and create triggers that may inspect 'channel history' via a message tool. Accessing chat/channel history and local files is consistent with a persistence layer but has privacy implications — the SKILL.md does not require unrelated system files or credentials.
ℹ
Install Mechanism
The registry contains no formal install spec (instruction-only), which is lower risk. However the docs recommend 'npm install -g @clawdactual/chitin' and running the chitin CLI; installing an external npm package will run third-party code on the host. The package and its postinstall scripts should be reviewed before running globally.
✓
Credentials
The skill declares no required environment variables or credentials. The SKILL.md references external sources (Carapace, message tools) but does not request unrelated secrets. This is proportionate to a local persistence tool, though any external integrations will require explicit credentials/configuration outside this document.
✓
Persistence & Privilege
The skill is not marked always:true and does not request persistent platform-level privileges in the metadata. Its intended persistence is a local database initialized by the CLI, which is appropriate for its purpose. There is no instruction to modify other skills or system-wide agent settings.
Assessment
This skill appears coherent with its stated purpose, but because it's instruction-only and encourages installing an npm package and reading channel history, take these precautions before installing: - Review the npm package (@clawdactual/chitin) source code and its GitHub repo (installing a global npm package runs third-party code). - Check where the CLI stores its database/files (default paths), what data it persists, and whether you’re comfortable storing conversation or behavioral data there. - Consider running the CLI in a sandbox or container and avoid -g/global install until you vet it. - Verify what external integrations (Carapace, message tool) require in terms of credentials and grant only least-privilege access. - If you have privacy concerns, audit or limit the triggers that cause the agent to read channel history or other message stores. If you can review the npm package source or ask the author for details about storage, encryption, and retention policies, that would raise confidence.

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

latestvk97fy2x9vjk7fdrc6dfn2dzzt9841mqm

License

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

Runtime requirements

🦞 Clawdis

Comments