ClawHub
Back to skill

Security audit

finding-startup-employees-for-recruiting

Security checks for vulnerabilities and agentic risk

Overview

This recruiting skill matches its stated purpose, but it should be reviewed because it collects identifiable candidate data through Apify without privacy guardrails and shows an unsafe API-token pattern.

Install only if you are comfortable sending recruiting targets and Twitter/X usernames to Apify. Confirm your use complies with platform terms, privacy rules, and employment laws, set tight limits on collection, avoid follower/following scraping unless necessary, minimize retention of candidate data, and use safer token handling such as authorization headers rather than token-bearing URLs.

A.I.G

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:88
Finding

API Token Exposed Through URL Query Strings

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 88 and 107
Vulnerability Type: Credential exposure through URL query parameters
Risk Level: Medium

Complete Code Snippets

bash
curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~tweet-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"searchTerms": ["at [COMPANY]", "engineer at [STARTUP]", "working at [SECTOR] startup"], "maxItems": 300}'
bash
curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~twitter-user-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"usernames": ["handle1", "handle2"]}'

Technical Analysis

The documented REST fallback places APIFY_TOKEN directly in the request URL. Although HTTPS encrypts the URL in transit, URL query strings are commonly retained by shell history, process or command telemetry, HTTP client diagnostics, monitoring platforms, reverse proxies, and server access logs.

When the shell expands $APIFY_TOKEN, these systems may record the resulting secret-bearing URL. Anyone with access to such records could recover the token and use it as a bearer credential. The issue affects both documented Apify actor calls.

Attack Path

  1. A user follows either REST fallback example with a valid APIFY_TOKEN.
  2. The shell expands $APIFY_TOKEN into the URL query string.
  3. The complete URL is captured by shell history, endpoint telemetry, debugging output, proxy logs, server logs, or copied diagnostic records.
  4. An attacker or unauthorized operator obtains access to one of those records.
  5. The attacker extracts the token and submits requests to the Apify API.
  6. The API accepts requests within the permissions and account scope granted to the compromised token.

Impact Assessment

Successful exploitation could allow unauthorized use of the victim's Apify account within the token's assigned privileges. Depe ...[truncated 358 chars]

Remediation
View remediation

Remediation Suggestions

  • Remove token=$APIFY_TOKEN from both request URLs.

  • Pass the token through Apify's supported authorization header rather than a query parameter. For example:

    bash
    curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~tweet-scraper/runs" \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"searchTerms": ["at [COMPANY]", "engineer at [STARTUP]", "working at [SECTOR] startup"], "maxItems": 300}'

  • Apply the same header-based authentication pattern to the twitter-user-scraper request.

  • Ensure command tracing and HTTP debug logging do not emit authorization headers.

  • Redact credentials from application, proxy, monitoring, and support logs.

  • If .env is used, exclude it from version control and restrict its filesystem permissions.

  • Rotate any token previously used in a query-string URL where logs or telemetry may have retained it.

  • Limit the replacement token to the minimum Apify permissions required for these actor runs.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The manifest says the skill returns structured fields including role level and career change signals, and the workflow checklist implies the skill will score candidate fit. However, the file contains only instructions to run Apify tweet/profile scrapers and then manually apply heuristic filtering/scoring from bio and tweet text; there is no implemented code or automated extraction logic shown for reliably producing those fields.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill is designed to collect, score, and export identifiable candidate data from Twitter/X for recruiting without any warning about privacy, lawful basis, retention, or platform terms. Even if the data is public, aggregating and exporting professional profiles at scale increases compliance and misuse risk, especially for targeted recruiting or competitor talent mapping.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The example input sets "tweetLanguage": "en", which imposes an English-language constraint in the workflow. The file does not offer a language choice or explain that the skill is intentionally limited to English, which is a natural-language policy concern under the locale rule.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

The skill explicitly sends search terms and associated scraping requests to a third-party service (Apify), which is an external data transmission event. In this context, the transmitted content may include recruiting targets, company names, and later usernames, which can expose sensitive business intent or collected personal data to an external processor.

Content

Scanner excerpt · SKILL.md (reported line 87)May include surrounding context.

REST API fallback:

bash
curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~tweet-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"searchTerms": ["at [COMPANY]", "engineer at [STARTUP]", "working at [SECTOR] startup"], "maxItems": 300}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 88)May include surrounding context.

REST API fallback:

bash
curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~tweet-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"searchTerms": ["at [COMPANY]", "engineer at [STARTUP]", "working at [SECTOR] startup"], "maxItems": 300}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

This step transmits Twitter usernames to Apify for enrichment, which is another concrete external data transfer involving identifiable individuals. In the recruiting context, bulk profile enrichment can reveal sourcing strategy and process personal data through a third party without any warning or consent/compliance guidance.

Content

Scanner excerpt · SKILL.md (reported line 107)May include surrounding context.

REST API fallback:

bash
curl -X POST \
  "https://api.apify.com/v2/acts/apidojo~twitter-user-scraper/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"usernames": ["handle1", "handle2"]}'

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The skill instructs users to use an APIFY_TOKEN but provides no warning about keeping it secret, avoiding command-history leakage, or preventing accidental inclusion in logs and shared files. API tokens grant access to external services and can be abused for unauthorized runs, data access, or billing consumption if exposed.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
73% confidence
Finding

The comment asserts that a helper script 'handles waiting, output, and file saving automatically,' which suggests an implemented automation layer. In this file, though, there is no code for that script—only example commands and manual API calls—so the documentation overstates what this skill file itself does.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.