plan-llm-cost-guardrails

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is strictly limited to an auditing and planning role. It performs diagnostic analysis of project source code to identify runaway cost risks and proposes remediation phases. All operations are local to the repository environment and do not involve unauthorized command execution or network exfiltration.\n- [INDIRECT_PROMPT_INJECTION]: As a tool that processes untrusted project source code to identify call sites and endpoints, the skill possesses an inherent attack surface for indirect prompt injection from the files it audits.\n
  • Ingestion points: The skill inventories and analyzes all LLM call sites and public AI endpoints within the project files.\n
  • Boundary markers: The instructions do not specify any boundary markers or ignore-rules for content found within the audited source code.\n
  • Capability inventory: The skill performs file system reads and text analysis to generate a markdown report (plan-llm-cost-guardrails.md).\n
  • Sanitization: No explicit sanitization or filtering of the audited code is defined in the procedure.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 09:31 PM