skills/canhassancode/skills/crucible/Gen Agent Trust Hub

crucible

Pass

Audited by Gen Agent Trust Hub on Nov 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes multiple shell commands to manage the testing environment, including git worktree add, git diff, timeout, and git worktree remove. It also executes a local shell script located at ../tester/scripts/check-locked-tests.sh.
  • [EXTERNAL_DOWNLOADS]: The skill uses npx stryker run in references/mutation.md, which may download the Stryker mutation testing tool from the npm registry if it is not present locally.
  • [DYNAMIC_EXECUTION]: In the 'Without Stryker' section of references/mutation.md, the agent is instructed to manually modify source code ('Hand-place mutants') by inverting conditions, dropping calls, or returning constants, and subsequently executing the test suite against these modifications.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from git diffs and ticket descriptions.
  • Ingestion points: Data enters via git diff output and the ticket ref argument.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the diff or ticket body are provided.
  • Capability inventory: The skill has the ability to execute shell commands, run external packages, and modify/execute local code.
  • Sanitization: There is no evidence of sanitization or validation of the content within the diff or ticket description before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 1, 2026, 10:15 AM