Analyze an authorized binary through the active Ghidra runtime and verify navigation in the mirrored GUI.
75%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Passed
No findings from the security scan
Fix and improve this skill with Tessl
tessl review fix ./packages/decepticon/decepticon/skills/standard/reverser/ghidra/SKILL.mdThe isolated E2B REA/Ghidra companion is the analysis runtime and the Run tab can mirror its CodeBrowser. The engagement agent discovers a binary under /workspace and opens it through rea_open_binary; the product pins its source digest and analysis run. Opening the viewer alone does not imply that a binary was analyzed.
Use rea_find_tools with a task-specific query to inspect the available native name and schema before calling it. REA provides a source-bound Evidence result and marks unresolved or derived observations. Do not infer tool availability from the full package catalog.
Address-specific calls can navigate the real Listing and Decompiler through the GUI mirror. Inspect the GUI result before saying the operator saw the location. If following failed, keep the REA analysis result and report the lack of visual confirmation separately. Record the binary hash, Evidence ID, and address with every finding.
A standalone run starts the optional reversing workload and uses rea_status, rea_find_tools(binary_path, query), and rea_call_tool(binary_path, name, arguments_json). The binary must be under the engagement's /workspace; no separate upload is needed. The sidecar pins the bytes and persists the original Evidence bundle under /workspace/.analysis/rea/<runId>/ before delivering a successful result. The complete available Ghidra-session tool set is pageable through rea_find_tools. Legacy headless tools remain a fallback when REA is unavailable; standalone output does not imply a hosted GUI update.
PurpleAILAB/Decepticon
49888ee
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.