skills/dnsqueen/orcinus-skills/build/Gen Agent Trust Hub

build

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes a 'calibrated brief' provided as external input to determine build slices and acceptance criteria. This represents an indirect prompt injection surface where malicious instructions embedded in the brief could influence the agent's actions during the implementation phase.
  • Ingestion points: The 'calibrated brief' is read in Step 1 to identify slices and acceptance criteria, and again in Step 3 to resolve design decisions.
  • Boundary markers: The instructions do not define specific delimiters or warnings to ignore potential injection content within the processed brief.
  • Capability inventory: The skill executes shell commands including build, lint, typecheck, test, and orcinus-verify as part of its verification loop.
  • Sanitization: No explicit sanitization, validation, or escaping steps are defined for the content ingested from the brief.
  • [COMMAND_EXECUTION]: The skill requires the agent to execute a series of shell commands as part of its core operational loop.
  • Evidence: Step 4 explicitly instructs running build, lint, typecheck, and test. Step 6 requires running the orcinus-verify command to validate the implementation before completion.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 01:45 PM