shorten
Verdict safe, but 1 finding (1 medium), mainly unauthorized tool use. Most recent scan 2026-02-07.
The skill has a MEDIUM severity issue due to missing implementation, creating a tool exploitation risk. The concept is straightforward (URL shortening via is.gd), but without actual scripts, the agent cannot safely execute this functionality and may be vulnerable to tool shadowing attacks. Additionally, there are LOW severity transparency issues: missing optional metadata fields and undeclared network access requirements. The skill would be safe once a proper implementation script is added with input validation and the metadata is completed to inform users about network dependencies.
Scanner's overall assessment (LLM analyzer)
Second scanner: NVIDIA SkillSpector
Independent static run, no language model, over the same files as the Cisco scan above. How the two scanners compare across the whole corpus: Two scanners, one corpus.
What this skill does
Shorten URLs using is.gd (no auth required). Returns a permanent short link.
- Package
- 3 files, 1.4 KB
- SKILL.md
- 50 words, 5 sections
- Code
- None. Instructions only.
SKILL.md covers: Shorten · Usage · Examples · Output: https://is.gd/O5d2Xq · Notes
Findings
MEDIUM Missing Implementation - Tool Exploitation Risk Unauthorized Tool Usellm
The skill declares functionality to shorten URLs using is.gd service but provides no implementation scripts. The SKILL.md shows bash usage examples ('shorten "https://example.com"') suggesting a bash command should exist, but no .sh or .py files are present. This creates a tool shadowing risk where the agent might attempt to execute a non-existent 'shorten' command, potentially allowing malicious actors to provide their own implementation or causing the agent to misinterpret the instruction and execute unintended commands.
Usage examples show: shorten "https://example.com/very/long/url" but no corresponding shorten.sh or shorten.py script exists in the packageWhat is in the package
| File | Kind | Size |
|---|---|---|
README.md | doc | 657 B |
SKILL.md | doc | 506 B |
package.json | config | 293 B |
External hosts referenced
is.gd github.com clawdhub.com google.com
Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.
Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.
What we checked
- Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
- Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
- Meta analyzer. A second model pass that validates, deduplicates and re-scores the other analyzers' findings.
What we did not check
- Runtime behaviour. The skill was not executed; findings come from reading its files.
- Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
- Content the analyzers cannot read: images, binaries, and some non-English text.
- Known scanner bypass classes. See the methodology and limitations page.
Sources
- Repository: https://clawhub.ai/skills/url-shortener
- Raw scan result: url-shortener-scan.json
- Badge for your README:
- Request a re-scan