Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Hubspot Audit
v1.0.0Run a comprehensive HubSpot CRM database audit. Analyzes contacts, companies, deals, engagement, data quality, and deliverability. Use when starting a CRM cl...
⭐ 0· 47·0 current·0 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.md clearly requires a HubSpot private app API token (HUBSPOT_API_TOKEN) and instructs creating a Python script to query HubSpot — this matches the stated purpose. However, the skill metadata declares no required environment variables or primary credential, which is inconsistent with needing the API token; that mismatch reduces trust and transparency.
Instruction Scope
Runtime instructions are focused on HubSpot queries, grading, and writing a markdown report (expected). They also instruct the agent to read/write a local .env file and persist the user's HUBSPOT_API_TOKEN to disk, and to create a scripts file and reports directory. Persisting secrets to .env is a scope decision the user should approve but is within the skill's functional needs.
Install Mechanism
There is no formal install spec in the registry, but the SKILL.md tells the operator to run `uv pip install hubspot-api-client python-dotenv`. 'uv' is not a standard installer wrapper and is not declared as a required binary — this is incoherent and potentially dangerous (typo or attempt to invoke an unexpected binary). The lack of an explicit, trusted install source increases risk.
Credentials
The skill legitimately needs one secret: a HubSpot API token. However, the registry metadata lists no required env vars or primary credential while the instructions explicitly rely on HUBSPOT_API_TOKEN. That mismatch is a transparency/least-privilege concern. No other unrelated credentials are requested.
Persistence & Privilege
The skill directs creating files (scripts/audit_portal.py, reports/*.md) and explicitly storing the provided HUBSPOT_API_TOKEN in a plaintext .env file. The skill is not marked 'always' and does not request special system privileges, but writing secrets to disk is persistent and increases risk if run in an untrusted environment.
What to consider before installing
This skill appears to implement a real HubSpot audit, but there are a few red flags to address before running it:
- The SKILL.md expects a HubSpot API token (HUBSPOT_API_TOKEN) but the skill metadata does not declare that credential — ask the publisher why the token is missing from the registry metadata.
- Do not run commands exactly as written until you confirm what 'uv pip install' is meant to be (likely a typo). Use a known installer (e.g., pip) or run dependency installation in an isolated environment (virtualenv, container).
- The instructions tell you to write the API token into a plaintext .env file. Prefer supplying the token via your platform's secret manager, environment variables injected at runtime, or a securely permissioned file rather than saving it in a repo or user home directory.
- Review the generated script (scripts/audit_portal.py) before executing it so you can verify it only calls the HubSpot APIs you expect and does not exfiltrate data elsewhere.
- If possible, create a HubSpot private app with the minimum required scopes (read-only where possible) and test the audit on a non-production portal or a limited dataset first.
If the publisher can confirm and fix the metadata to declare HUBSPOT_API_TOKEN, correct the install instruction to a standard command (or provide an explicit install spec), and document where secrets are stored, the skill would be much more acceptable.Like a lobster shell, security has layers — review code before you run it.
latestvk9750by0gjgg1wrf1gpv3a1t8h83nt0z
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
