ClawHub

Genai Toolkit

v1.0.0

Bridge AI models to databases through MCP with config and evaluation tools. Use when setting up DB tools, comparing engines, or evaluating prompt quality.

⭐ 0· 85·0 current·0 all-time
bybytesagain4@xueyetianya
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 promise is a CLI toolkit for logging and evaluating generative-AI experiments. The included shell script implements the described commands and stores data under ~/.local/share/genai-toolkit, so required capabilities match the stated purpose.
✓
Instruction Scope
SKILL.md instructs only local logging, exporting, searching, and status checks. The script reads/writes only within the data directory and standard system utilities (date, grep, tail, wc, du). There are no instructions to read unrelated system files, environment secrets, or to transmit data externally.
✓
Install Mechanism
No install spec is provided (instruction-only skill). A single bash script is included; it is a plain shell implementation and does not download or extract remote code. Risk from install mechanism is low.
ℹ
Credentials
The skill requests no environment variables or credentials. It does use $HOME to create ~/.local/share/genai-toolkit (expected). SKILL.md suggests logging items like 'API keys or environment settings' — this is a usage note rather than an automatic collection, but users should avoid storing secrets in plaintext log files.
✓
Persistence & Privilege
Skill is not marked always:true and does not modify other skills or system-wide settings. It persists only its own data under the user's home directory and does not request elevated privileges.
Assessment
This appears to be a simple local logging CLI. Before installing: (1) Review the script if you want to be certain it won’t be changed; (2) be aware it stores all entries in plaintext at ~/.local/share/genai-toolkit — do not log secrets (API keys, passwords) there; (3) consider setting restrictive permissions on that directory (chmod 700) or using an encrypted workspace if you must record sensitive data; (4) exports (json/csv/txt) are created as files and could be shared accidentally, so handle exported files carefully.

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

latestvk97apc4gb20psrq7a9vvmynhhh8367j7

License

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

Comments