agile-v-aibom

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external AI agent runs, including runtime logs, session metadata, and CI hooks. This creates a surface for indirect prompt injection if malicious instructions are embedded in the logs being analyzed.\n
  • Ingestion points: SKILL.md specifies capturing 'declared AI context' and 'observed AI context' from runtime logs, CI hooks, and session metadata.\n
  • Boundary markers: Relies on normalization into structural formats (YAML/JSON) and validation against predefined templates and schemas referenced in SKILL.md.\n
  • Capability inventory: Capabilities are limited to reading logs, writing metadata files, and performing manifest comparisons; no direct execution of ingested content is specified.\n
  • Sanitization: Safety rules explicitly require the redaction of secrets, credentials, and PII before inclusion in generated manifests.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 06:07 AM
Security Audit — agent-trust-hub — agile-v-aibom