Skill flagged — suspicious patterns detected

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

BotLearn Assessment

v1.0.1

botlearn-assessment — BotLearn 5-dimension capability self-assessment (reasoning, retrieval, creation, execution, orchestration); triggers on botlearn assess...

0· 48·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
Name/description match the visible behavior: an autonomous 5-dimension self-assessment. However, the SKILL.md expects flows, question banks, and a node script to exist (e.g., flows/* files, scripts/radar-chart.js) although the package only includes three markdown files. The skill does not declare required binaries (node) or any install steps even though it instructs running a Node command and writing multiple files. This mismatch is disproportionate to the stated purpose.
!
Instruction Scope
Instructions demand fully autonomous behavior (do not ask user for anything), scanning for available tools, skipping questions if capabilities/tools are missing, writing multiple report files under results/, and running node scripts. The skill also instructs to always output full file paths for generated reports. It references many flow files (flows/exam-execution.md, flows/full-exam.md, flows/generate-report.md, etc.) that are not present in the bundle — the agent will attempt to read/require them and then skip or behave unpredictably. Asking the agent to 'search for available tool or installed skill' may cause it to enumerate system/tooling state.
Install Mechanism
There is no install spec (instruction-only), which is low-risk in general. However, runtime instructions call node scripts (node scripts/radar-chart.js) and reference generated artifacts; requiring node (or other runtime dependencies) is not declared. That mismatch increases operational risk: the skill expects execution capabilities that are not promised or provided.
Credentials
The skill declares no environment variables or credentials (ok). Still, it instructs writing files (results/...), updating INDEX.md, and printing full file paths — this can expose filesystem layout and persistent artefacts. While not requesting secrets, the file-write behavior and the expectation to run local scripts should be reviewed before installation.
Persistence & Privilege
always is false and the skill does not request to modify other skills or global agent settings. It does request creation of persistent report files in a results/ directory and to update INDEX.md; this is typical for a reporting tool but should be accepted only if you trust where files will be stored.
What to consider before installing
What to check before installing or enabling this skill: - Confirm the missing files and scripts: SKILL.md references many flow files and a Node script (scripts/radar-chart.js) that are not bundled. Ask the publisher or inspect the full skill package to ensure those files exist and are safe. - Verify runtime requirements: The skill expects to run `node` (and possibly other tooling). If you permit it to run, ensure Node.js and the specific scripts are available and reviewed. - Review file-write behavior: The skill will write persistent reports into a results/ directory and update an INDEX.md, and it will provide full filesystem paths. If that is sensitive in your environment, run the skill in a sandbox or deny write access. - Autonomous behavior: The skill enforces 'do not ask user' and auto-start. If you do not want an autonomous agent to begin exams and create files without confirmation, do not grant it broad runtime privileges or enable it in always-on contexts. - Test in isolation: If you want to try it, run it in an isolated environment (container or VM) after verifying the missing scripts; confirm the node script and other referenced flow files are present and safe. Additional information that would change this assessment: inclusion of the referenced flows and scripts in the package, an explicit declaration of required binaries (e.g., NODE), and a trusted publisher/homepage. If those exist and are legitimate, confidence would increase toward benign.

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

latestvk9725yv4yb105rzfsrgcdq1jvx83g6m9

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments