skills/sahin/claude-skills/tech-debt/Gen Agent Trust Hub

tech-debt

Pass

Audited by Gen Agent Trust Hub on Apr 1, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests data from several local JSON files, establishing an indirect prompt injection surface if those files contain data from untrusted sources (e.g., automated PR comments or external incident reports).\n
  • Ingestion points: data/engineering/tech_debt.json, data/engineering/tech_stack.json, data/engineering/engineering_scorecard.json, and data/cfo/latest_forecast.json (defined in SKILL.md).\n
  • Boundary markers: None present to distinguish data from instructions.\n
  • Capability inventory: The skill has the capability to write to the data/engineering/tech_debt.json file.\n
  • Sanitization: No validation or sanitization of the ingested data is performed.- [SAFE]: The skill's file access is limited to its own data directory and the requested files are relevant to its primary function as a technical debt analyst.\n
  • Evidence: Reading data/cfo/latest_forecast.json is a standard business context requirement for the stated purpose of impact-based prioritization.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 1, 2026, 12:24 AM
Security Audit — agent-trust-hub — tech-debt