pain-language-engagers
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The Python script
pain_language_engagers.pycommunicates with the Apify API (api.apify.com) to trigger scraping actors and retrieve datasets. These actors includeharvestapi/linkedin-post-search,harvestapi/linkedin-company-posts, andsupreme_coder/linkedin-profile-scraper.\n- [DATA_EXFILTRATION]: The skill extracts public LinkedIn profile information (name, headline, location) and engagement data (post URLs, reactions, and comment text). This data is aggregated and written to local CSV files in theoutput/directory.\n- [COMMAND_EXECUTION]: The skill requires the execution of a Python script using the command line (e.g.,python3 skills/pain-language-engagers/scripts/pain_language_engagers.py). This script performs network operations and local file writes.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from LinkedIn.\n - Ingestion points: LinkedIn comment content and post text are fetched via Apify actors and processed in
scripts/pain_language_engagers.py.\n - Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious content embedded within the scraped LinkedIn comments.\n
- Capability inventory: The script has the ability to make network requests (to Apify) and write files locally. The agent summarizing the results could be misled by injection attempts in the scraped data.\n
- Sanitization: The script does not sanitize or filter the content of the LinkedIn comments before including them in the final report and presenting them to the user.
Audit Metadata