High-risk tools in Cortex
7 of the 32 tools in Cortex are classified as high risk. This page profiles those tools specifically, with recommended policy actions and the attack patterns that target them.
Every operation listed below is an action PolicyLayer recommends controlling at the transport layer. Open any tool to see the full profile, risk score, and YAML policy snippet.
Tools at high risk
-
builderExecuteBuild profile. May edit source and test files but not config,
-
context.reviewExecuteRun enterprise policy validators against the current project.
-
cortex.workflow.advanceExecuteComplete the current stage of a workflow run by writing its artifact and advancing the run pointer. Returns the new run state plus the next stage
-
cortex.workflow.envelopeExecuteCompose the prompt envelope for a workflow stage without advancing the run. Defaults to the run
-
cortex.workflow.startExecuteStart a Cortex Harness workflow run for a task. Creates .agents/<task_id>/state.json and returns the first stage
-
cortex.workflow.statusExecuteRead the current run state for a task (current stage, completed stages, outcome). Returns null state when no run exists for the given task_id. Enterprise-only.
-
policy.syncExecuteTrigger manual policy sync. Connected: pulls from cloud API. Air-gapped: reloads local org-rules.yaml.
Attacks that target this class
High-risk tools in any server share these documented attack patterns. Each links to the full case and the defensive policy.