last30days
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- PROMPT_INJECTION (LOW): Potential for indirect prompt injection from untrusted external data. \n
- Ingestion points:
scripts/lib/reddit_enrich.py(crawls Reddit comment bodies) andscripts/lib/xai_x.py(crawls X post text). \n - Boundary markers: Absent; ingested content is rendered into the final context report without explicit delimiters or instructions to ignore embedded commands. \n
- Capability inventory: The skill is restricted to standard HTTP requests to whitelisted providers and local file operations; it does not possess dangerous capabilities like arbitrary code execution or shell access. \n
- Sanitization: Data is truncated to token limits, but the skill does not sanitize the text for malicious instructions targeting the reading agent. \n- DATA_EXFILTRATION (SAFE): Sensitive API keys are managed via a local config file (
~/.config/last30days/.env) and are only transmitted to authorized OpenAI and xAI endpoints. \n- EXTERNAL_DOWNLOADS (SAFE): Network operations are restricted toapi.openai.com,api.x.ai, andreddit.comusing standard libraries.
Audit Metadata