atlassian-templates
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill documentation and reference files (e.g., confluence-templates.md, jira-templates.md) contain standard, legitimate project management templates. No prompt injection or malicious instructions were detected.\n- [SAFE]: The Python utilities in the scripts/ directory are benign automation tools using standard library modules (argparse, json, sys, datetime). They do not include any dangerous functions like eval() or perform network-based exfiltration.\n- [INDIRECT_PROMPT_INJECTION]: The Python scripts process external JSON files, creating a surface for indirect prompt injection if the ingested data contains instructions targeting the agent. However, the capabilities are low-risk and the behavior is typical for data-processing tools.\n
- Ingestion points: scripts/template_scaffolder.py, scripts/template_usage_analyzer.py, and scripts/template_validator.py read JSON files from user-specified paths.\n
- Boundary markers: None present; the scripts interpolate JSON values directly into templates or reports without delimiters.\n
- Capability inventory: The scripts perform local file reading and print results to stdout. They do not invoke subprocesses, write to files, or make network calls.\n
- Sanitization: The scripts use json.load() for safe parsing, but no further sanitization is applied to the string content within the JSON fields.
Audit Metadata