asset-spec
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_CONTEXT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to execute a local scriptyaml-helper.shlocated in the project's hooks directory. This is used to resolve configuration settings like review modes and automation levels.\n- [DYNAMIC_CONTEXT_INJECTION]: A shell command is executed at skill load time using the!syntax to initialize configuration. This involves calling theyaml-helper.shscript with specific flags.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from various project files (GDDs, manifests, character profiles) which is then used as context for sub-agents (art-director,technical-artist). This creates an attack surface if those files contain adversarial instructions.\n - Ingestion points:
design/gdd/*.md,design/narrative/characters/*.md,design/assets/entity-inventory.md,design/art/art-bible.md.\n - Boundary markers: Not explicitly defined in the sub-agent prompt instructions.\n
- Capability inventory:
Bash,Write,Edit,Agent,AskUserQuestion.\n - Sanitization: None detected in the provided instructions.
Audit Metadata