stock-evaluator-v3

SAFE max severity MEDIUMSkillSpector Caution 8 registry clawhub.ai last scanned 2026-02-07 1 scan

Verdict safe, but 2 findings (2 medium), mainly unauthorized tool use, resource abuse. Most recent scan 2026-02-07.

The stock-evaluator-v3 skill is a comprehensive stock analysis tool with MODERATE security concerns. The skill demonstrates good security practices by mandating web searches with cited sources and explicitly prohibiting data fabrication. However, it has transparency issues (missing allowed-tools declaration), potential resource exhaustion risks (unbounded web searches without rate limiting), and package integrity concerns (19 referenced files not found). The skill does NOT exhibit high-risk behaviors like data exfiltration, prompt injection, credential theft, or command injection. The primary concerns are operational robustness and transparency rather than malicious intent. With the recommended remediations (adding allowed-tools, implementing rate limiting, and resolving missing file references), this skill would be suitable for use.
Scanner's overall assessment (LLM analyzer)

Findings2

Critical0

High0

Medium2

Low0

Analyzersstatic llm meta

Second scanner: NVIDIA SkillSpector

VerdictCaution 8

Score8 / 100

Issues1

Categoriesanalysis-evasion

Version2.12.0 (static)

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

Comprehensive evaluation of potential stock investments combining valuation analysis, fundamental research, technical assessment, and clear buy/hold/sell recommendations. Use when the user asks about buying a stock, evaluating investment opportunities, analyzing watchlist candidates, or requests stock recommendations. Provides specific entry prices, position sizing, and conviction ratings.

Package: 5 files, 187 KB
SKILL.md: 12507 words, 12+ sections
Code: None. Instructions only.

SKILL.md covers: Stock Evaluator (Enhanced) · ⚠️ CRITICAL: MANDATORY DELIVERABLES CHECKLIST · ⚠️ CRITICAL: DATA INTEGRITY RULES · ZERO FABRICATION POLICY · MANDATORY WEB SEARCHES (minimum per analysis) · DATA SOURCE HIERARCHY · PROHIBITED · HANDLING UNAVAILABLE DATA · STANDARDIZED METRIC LABELS · Row 1: PRICE & VALUATION | FINANCIAL PERFORMANCE · Row 2: GROWTH METRICS | RISK INDICATORS · Row 3: LIQUIDITY & FREE CASH FLOW | INSIDER & SENTIMENT & CLASS

Findings

MEDIUM Missing allowed-tools Declaration Unauthorized Tool Usellm

The skill manifest does not specify the 'allowed-tools' field, which is optional but recommended for transparency. Given the skill's requirements for web searches and data retrieval, users cannot verify which agent tools (Read, Write, Bash, Python, etc.) the skill intends to use. This reduces transparency about the skill's capabilities and potential system access.

SKILL.md:1

YAML Manifest shows: allowed-tools: Not specified

Remediation: Add an 'allowed-tools' field to the YAML frontmatter specifying which agent tools are required (e.g., [Read, Bash] for web searches, or [Read, Python] if using Python for data processing).

MEDIUM Unbounded Web Search Requirements Without Rate Limiting Resource Abusellm

The skill mandates a minimum of 8 web searches per analysis with no rate limiting, timeout controls, or failure handling specified. The instruction 'If you cannot complete any item, STOP and ask for clarification' could lead to analysis paralysis if searches fail. No guidance on retry limits, timeout thresholds, or graceful degradation when data sources are unavailable.

SKILL.md

MANDATORY WEB SEARCHES (minimum per analysis) - lists 8 required searches with no timeout/retry limits specified. Instructions state 'NEVER fabricate' but don't handle search failures gracefully.

Remediation: Add explicit guidance for: (1) Maximum retry attempts per search (e.g., 3 retries), (2) Timeout thresholds (e.g., 30 seconds per search), (3) Graceful degradation when searches fail (allow partial analysis with clear disclosure), (4) Rate limiting to prevent resource exhaustion.

What is in the package

File	Kind	Size
`ADVANCED-METRICS-GUIDE.md`	doc	33 KB
`DASHBOARD-GUIDE.md`	doc	13 KB
`EVALUATION-WORKFLOWS.md`	doc	25 KB
`SKILL.md`	doc	94 KB
`VALUATION-GUIDE.md`	doc	22 KB

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/stock-evaluator
Raw scan result: stock-evaluator-scan.json
Badge for your README: ![skillscan.io](https://skillscan.io/badge/clawhub-ai-stock-evaluator-v3.svg)
Request a re-scan