Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
BS Deep Analysis
v1.0.2Controller-level Balance Sheet deep analysis from QuickBooks Online. Pulls current and prior period BS, runs 3-month rolling averages, GL drill-down for mate...
⭐ 0· 112·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
high confidencePurpose & Capability
The skill's stated purpose is to pull data from QuickBooks Online (QBO) and run a local Python pipeline. However, the package declares no required credentials, no primary credential, and no config paths even though the SKILL.md repeatedly references a 'qbo-client' connection and a sandbox flag. That is inconsistent: a QBO integration normally requires API keys/tokens or a configured client. Also the SKILL.md references a local script path (scripts/pipelines/bs-deep-analysis.py) that is not included in the skill bundle.
Instruction Scope
Instructions direct running a local Python script, accessing QBO to pull multiple reports (BS snapshots, GL, P&L), and writing cache and Excel outputs to disk. The instructions also reference a configurable limit (GL_MAX_ROWS_PER_ACCOUNT). The skill does not instruct exfiltration to external endpoints, but it does assume access to QBO credentials and local filesystem paths (.cache and output dirs). The combination of assuming privileged access (QBO) and missing authentication details is a scope/information gap that could lead users to supply credentials without knowing where code runs.
Install Mechanism
This is an instruction-only skill with no install spec and no included code files. That lowers installation risk (nothing is downloaded or written by an install step). However, the SKILL.md expects a local script to exist; since it's not present, users would need to obtain that code from elsewhere — which is an operational/verification risk but not an install-time code injection in this bundle.
Credentials
No environment variables or credentials are declared in the registry metadata, yet the run instructions require a 'qbo-client' connection and mention a configurable GL_MAX_ROWS_PER_ACCOUNT. A QBO integration is expected to need API credentials or an OAuth client — asking for those implicitly while declaring none is disproportionate and opaque. Users could be prompted to provide sensitive QuickBooks credentials without clarity on where/how they're stored or used.
Persistence & Privilege
The skill does not request always:true and does not modify other skills. It writes a cache file under .cache/bs-deep-analysis/{slug}.json and outputs Excel files to a user directory by default, which is consistent with its stated CDC/cache behavior. Persisting run caches is expected, but users should be aware these caches will contain financial data on disk.
Scan Findings in Context
[no_code_files_found] unexpected: The static scanner found no code files because this is an instruction-only skill, but the SKILL.md points to a local Python script (scripts/pipelines/bs-deep-analysis.py) that is not included. For a skill that instructs running code that accesses QBO, the absence of that code is unexpected and concerning.
What to consider before installing
Do not run or supply credentials to this skill yet. The SKILL.md expects a local pipeline script and a connected 'qbo-client' but the skill bundle contains no code and declares no QBO credentials—this is an incoherence. Before installing or using: 1) Ask the publisher for the source repository or the exact script referenced (verify its code and origin). 2) Confirm what authentication method is used for QBO (OAuth client, API key, service account) and where credentials are stored; never paste secrets into an unverified interface. 3) Verify how GL_MAX_ROWS_PER_ACCOUNT (and any other config) is set and whether any sensitive data is transmitted off your machine. 4) Run the pipeline in a controlled environment (sandbox VM) first and inspect .cache files and generated Excel output for unexpected network calls or telemetry. If the publisher cannot provide a trustworthy code repo and explicit credential requirements, treat the skill as untrusted.Like a lobster shell, security has layers — review code before you run it.
latestvk973nk8yxkfdvk4y29yhz2mnj983c0zs
License
MIT-0
Free to use, modify, and redistribute. No attribution required.