llm
Verdict safe, but 2 findings (2 medium), mainly data exfiltration, resource abuse. Most recent scan 2026-02-06.
This skill presents MODERATE security concerns primarily due to lack of implementation transparency. While the concept of multi-provider LLM integration is legitimate, the absence of any Python or Bash scripts makes it impossible to verify critical security properties: API credential handling, data transmission security, input sanitization, and prevention of data exfiltration. The 'always: True' automatic activation flag combined with external API dependencies creates risk of unintended API calls and cost accumulation. The skill requires external binaries (curl, jq) which could be vectors for command injection if user input is not properly sanitized. Missing metadata (license, compatibility, allowed-tools) reduces transparency. The skill would benefit from: (1) providing implementation scripts for security review, (2) documenting credential management practices, (3) reconsidering automatic activation, (4) adding proper metadata fields.
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
Multi-provider LLM integration. Unified interface for OpenAI, Anthropic, Google, and local models.
- Package
- 1 file, 644 B
- SKILL.md
- 66 words, 4 sections
- Code
- None. Instructions only.
- Requires
- bins:
curljq
SKILL.md covers: LLM 🔮 · Supported Providers · Features · Usage Examples
Findings
MEDIUM External API Integration Without Implementation Details Data Exfiltrationllm
The skill claims to integrate with multiple external LLM providers (OpenAI, Anthropic, Google, local models) but provides no implementation scripts. This creates uncertainty about how API credentials are handled, whether data is transmitted securely, and if there are any hardcoded secrets. The 'always: True' flag in clawdbot metadata suggests automatic activation, which could lead to unintended API calls.
SKILL.md
Supported Providers: OpenAI, Anthropic, Google, Ollama, LM Studio. Metadata shows 'always: True' but no Python/Bash scripts provided.MEDIUM Automatic Activation Without User Control Resource Abusellm
The clawdbot metadata specifies 'always: True', meaning this skill is automatically active without explicit user invocation. Combined with external API integrations, this could lead to unintended API calls, token consumption, or cost accumulation without user awareness or consent.
SKILL.md
clawdbot: {'emoji': '🔮', 'always': True, 'requires': {'bins': ['curl', 'jq']}}What is in the package
| File | Kind | Size |
|---|---|---|
SKILL.md | doc | 644 B |
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/llm
- Raw scan result: llm-scan.json
- Badge for your README:
- Request a re-scan