github.com/datadog/system-tests · main · scanned 2 h ago · 22 config files

22 agent components (1 MCP server), declared 23 times in datadog/system-tests: 2 unvetted.

Components221 MCP server
High blast radius0credentials, shell or unvetted code
Credentials handed to agents0none seen
Not vetted2of 2 local servers, skills and plugins · 1 hosted
ComponentWhereVerdictBlast radius
AGENTS.mdInstructions · any agent
AGENTS.mdclean
low
  • is read by the agent as instructions on every session
aws-ssi-testing.mdcInstructions · Cursor
.cursor/rules/aws-ssi-testing.mdcclean
low
  • is read by the agent as instructions on every session
CLAUDE.mdInstructions · any agent
CLAUDE.mdclean
low
  • is read by the agent as instructions on every session
code-format-standards.mdcInstructions · Cursor
.cursor/rules/code-format-standards.mdcclean
low
  • is read by the agent as instructions on every session
datadogMCP server · Cursor · httpmcp.datadoghq.com
.cursor/mcp.jsonhosted · no package
low
  • sends the agent's tool calls and their data to mcp.datadoghq.com
  • is not a registry package, so there is no verdict to check
datadogMCP server · Claude Code · httpmcp.datadoghq.com
.mcp.jsonhosted · no package
low
  • sends the agent's tool calls and their data to mcp.datadoghq.com
  • is not a registry package, so there is no verdict to check
ddPlugin · Claude Code
.claude/settings.jsonnot vetted
low
  • adds instructions and scripts the agent will follow
  • has no KillerSkills verdict yet
devtools.mdcInstructions · Cursor
.cursor/rules/devtools.mdcclean
low
  • is read by the agent as instructions on every session
doc.mdcInstructions · Cursor
.cursor/rules/doc.mdcclean
low
  • is read by the agent as instructions on every session
end-to-end-testing.mdcInstructions · Cursor
.cursor/rules/end-to-end-testing.mdcclean
low
  • is read by the agent as instructions on every session
fine-tuning-guidance.mdcInstructions · Cursor
.cursor/rules/fine-tuning-guidance.mdcclean
low
  • is read by the agent as instructions on every session
general-behavior.mdcInstructions · Cursor
.cursor/rules/general-behavior.mdcclean
low
  • is read by the agent as instructions on every session
java-endpoint-prompt.mdcInstructions · Cursor
.cursor/rules/java-endpoint-prompt.mdcclean
low
  • is read by the agent as instructions on every session
k8s-ssi.mdcInstructions · Cursor
.cursor/rules/k8s-ssi.mdcclean
low
  • is read by the agent as instructions on every session
marketplace-auto-updatePlugin · Claude Code
.claude/settings.jsonnot vetted
low
  • adds instructions and scripts the agent will follow
  • has no KillerSkills verdict yet
parametric-testing.mdcInstructions · Cursor
.cursor/rules/parametric-testing.mdcclean
low
  • is read by the agent as instructions on every session
pr-review.mdcInstructions · Cursor
.cursor/rules/pr-review.mdcclean
low
  • is read by the agent as instructions on every session
promptfoo-llm.mdcInstructions · Cursor
.cursor/rules/promptfoo-llm.mdcclean
low
  • is read by the agent as instructions on every session
repository-structure.mdcInstructions · Cursor
.cursor/rules/repository-structure.mdcclean
low
  • is read by the agent as instructions on every session
system-tests-overview.mdcInstructions · Cursor
.cursor/rules/system-tests-overview.mdcclean
low
  • is read by the agent as instructions on every session
tag-official-workflows.mdcInstructions · Cursor
.cursor/rules/tag-official-workflows.mdcclean
low
  • is read by the agent as instructions on every session
test_the_test.mdcInstructions · Cursor
.cursor/rules/test_the_test.mdcclean
low
  • is read by the agent as instructions on every session
test-activation.mdcInstructions · Cursor
.cursor/rules/test-activation.mdcclean
low
  • is read by the agent as instructions on every session
What to do about it
  • Hosted servers have no package to run in the sandbox. Everything the agent sends them goes to that host, so treat the credential as the trust decision. Teams can watch a hosted server's tool list for changes.
  • Unvetted means we have no sandbox run for that exact package. Ask your agent to check before installing: claude mcp add killerskills -- npx -y vet-skills mcp
  • For a whole organization (every repo and laptop, rescanned daily, with removal across machines), see Discovery for teams.
Share: https://killerskills.ai/scan/github/datadog/system-testsRescan nowScan another