reverse-engineer-anything

1# REA
2 
3Use REA when a claim depends on a shipped binary or package, decompilation,
4requested runtime evidence, or comparison with
5behavior not established by available source. For ordinary analysis of a
6complete source repository, use normal repository tools and do not run REA
7readiness or provider commands.
8 
9## Connect only when needed
10 
11Installing this skill supplies instructions; it does not register the REA MCP
12server, install analysis engines, or add tools to an already-running agent.
13The exception is a Claude Code, Codex, or Cursor Marketplace plugin install:
14the plugin bundles this skill and registers its own `rea` server, outside the
15state that `rea setup` and `rea doctor` manage.
16 
17If REA tools are available and their registration is not known to be stale,
18proceed directly to the target. Do not run diagnostics or setup before every
19investigation. Use the connected server's actual tool list and input schemas;
20a skill installed from repository main may describe capabilities absent from
21an older npm release. Keep complete inline Evidence when that server does not
22advertise retained references.
23 
24If tools are absent, a call is blocked by the client, or a required provider is
25unavailable, read [connection and recovery](references/connection-and-recovery.md).
26Diagnose the observed blocker, continue useful supported work, and resume the
27original task after repair. Client approval, shell PATH, registration and engine
28…