Headless Ghidra Intake

Headless Ghidra Intake — P0

P0 prepares a target for the rest of the workflow. It is responsible for turning an input binary or archive path into a valid ghidra-agent-cli workspace target and recording the initial analysis scope.

Required ghidra-agent-cli Commands

• ghidra-agent-cli workspace init
• ghidra-agent-cli ghidra discover
• ghidra-agent-cli inspect binary
• ghidra-agent-cli scope show
• ghidra-agent-cli scope set
• ghidra-agent-cli scope add-entry
• ghidra-agent-cli scope remove-entry
• ghidra-agent-cli gate check --phase P0

Archive normalization and any target-specific bootstrap scripts may still run as backend details until the CLI exposes them directly.

Inputs

• User-provided binary or archive path
• Workspace root
• Local environment needed for Ghidra discovery

Outputs

• targets/<target-id>/ghidra-projects/
• artifacts/<target-id>/pipeline-state.yaml
• artifacts/<target-id>/scope.yaml
• Optional phase-owned intake records under artifacts/<target-id>/intake/

Exit Expectations

• Target workspace exists and is addressable through --target <id>.
• pipeline-state.yaml records the selected binary path.
• scope.yaml records the selected functions, addresses, symbols, or explicit whole-target scope.
• Ghidra discovery and binary inspection have been run and reviewed.

Constraints

• Do not run actual Ghidra analysis in this phase.
• Do not write baseline, evidence, decompilation, or verification outputs.
• Do not bypass ghidra-agent-cli workspace init, ghidra discover, or inspect binary for supported steps.
• Do not create or run a new Ghidra script if the CLI lacks a capability; pause and ask the user first.

Next Step

• P0 gate passes → headless-ghidra-baseline