diff-summary
Verdict safe, but 3 findings (3 medium), mainly social engineering, data exfiltration, unauthorized tool use. Most recent scan 2026-02-06.
This package has MEDIUM severity security concerns. The primary issue is that it's not actually an agent skill - it's documentation for an external NPM CLI tool (npx ai-diff-summary) masquerading as an agent skill package. The SKILL.md contains no agent instructions, only CLI usage documentation. Additionally, it requires an OPENAI_API_KEY without security guidance, references a missing main.py file, and lacks proper metadata. The package appears to be either incorrectly packaged or intentionally deceptive. It should either be converted to a proper agent skill with bundled scripts and agent instructions, or clearly labeled as external tool documentation.
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
Summarize git diffs in plain English
- Package
- 1 file, 1.2 KB
- SKILL.md
- 159 words, 12+ sections
- Code
- None. Instructions only.
SKILL.md covers: Diff Summary · Quick Start · What It Does · Usage Examples · Summarize staged changes · Summarize specific commit · Compare branches · Output as PR description · Output Example · Summary · Changes · Breaking Changes
Findings
MEDIUM Misleading Skill Package - Not an Agent Skill Social Engineeringllm
This package is advertised as an agent skill but is actually an NPM package (npx ai-diff-summary) that requires external installation and execution. The SKILL.md contains no agent instructions, only documentation for a standalone CLI tool. This is deceptive packaging that misleads users about the skill's nature and functionality.
SKILL.md:1
Quick Start section shows 'npx ai-diff-summary' - this is an NPM package installation command, not agent skill usage. No agent instructions like 'When the user asks to...' are present.MEDIUM External API Key Requirement - Credential Exposure Risk Data Exfiltrationllm
The skill requires OPENAI_API_KEY environment variable but provides no guidance on secure handling. Users may expose their API keys through environment variables that could be accessed by other processes or logged. The skill does not specify how the key is used or transmitted.
SKILL.md:45
'OPENAI_API_KEY required' - No security guidance provided for API key handling, storage, or transmission.MEDIUM Missing Script Implementation - Broken Functionality Unauthorized Tool Usellm
The instructions reference 'main.py' as a script file, but this file is not found in the package. This indicates either incomplete packaging or a tool shadowing attempt where the actual implementation is hidden or replaced by the external NPM package.
SKILL.md:1
Referenced file 'main.py' not found in package. Instructions show NPX commands instead of Python execution.What is in the package
| File | Kind | Size |
|---|---|---|
SKILL.md | doc | 1.2 KB |
External hosts referenced
github.com x.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.
- Behavioral analyzer. Checks what the skill asks the agent to do against what its description claims it does.
- Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
- Trigger analyzer. Looks for instructions that fire on conditions the user did not ask for.
- 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/ai-diff-summary
- Raw scan result: git-diff-summarizer-scan.json
- Badge for your README:
- Request a re-scan