Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@re-binary-diff diff original.bin and patched.bin"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server for read-only binary comparison: a unified diff between two files, and a per-section fingerprint of one file. Dry-run only — the server never writes a byte to disk.
The 2026-06-05 stress test surfaced a need to compare an
original binary against a patched copy (the Output/.../patches/
workflow) without re-introducing the on-disk patch primitive.
re-binary-diff is the read-only cousin: it reports the diff,
it never applies it.
Related MCP server: BinaryAnalysis-MCP
Tool | What it does |
| Health check — |
| Run |
| Return per-chunk SHA-256 + offset + size for a single file (a structural fingerprint, like |
Part of the RE-AI plugin; ./install.sh installs the package. To
install standalone:
pip install -e ./servers/re-binary-diffre-binary-diff # stdio transport (default for MCP)
python -m re_binary_diff # equivalentReturn server status + version. Always status: OK —
re-binary-diff has no external system dependencies.
Inputs: no named parameters.
Return a per-chunk SHA-256 + offset + size fingerprint of path.
Useful for finding the structural delta between two
versions of a binary without loading either into
memory. Pair with unified_diff (chunk mode) for
large files.
Args: path: file to fingerprint chunk_size: bytes per chunk (default 1 MiB) max_chunks: 0 = no cap; > 0 = return at most N chunks
Returns::
{
"path": "...",
"size": N,
"chunk_size": N,
"truncated": bool,
"chunks": [
{"index": 0, "offset": 0, "size": 1048576, "sha256": "..."},
...
],
}
Inputs: path (required), chunk_size, max_chunks.
Compute a unified-style diff between path_a and path_b.
For small files (<= 8 MiB per side) the function runs
difflib.unified_diff over the byte streams (decoded
as latin-1). For large files it falls back to a
per-chunk SHA-256 diff.
Args: path_a: first file (the "before" or "original") path_b: second file (the "after" or "patched")
Returns::
{
"path_a": "...",
"path_b": "...",
"size_a": N,
"size_b": N,
"mode": "sha256_equal" | "inline_unified_diff" | "chunk_sha256_diff",
"equal": bool,
"diff_lines": ["...", ...], (inline mode)
"chunk_diff": [{...}, ...], (chunk mode)
}The function is read-only. The "patched" copy at path_b is not modified.
Inputs: path_a (required), path_b (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 3 tools
Each tool has a clearly distinct purpose: a health check, a single-file fingerprint, and a two-file diff. There is no ambiguity about which tool to use for a given task.
Tool names use different patterns: 'check_binary_diff' (verb_noun), 'fingerprint_sections' (verb_noun), and 'unified_diff' (adjective_noun). The mixing of conventions reduces consistency.
With 3 tools, the server is well-scoped for binary diffing tasks. It covers the essential operations without excess, though some might argue a tool to list chunks could be merged.
The tool set covers the core workflow: fingerprint a file, compare two files, and check server status. There are minor gaps (e.g., no explicit tool to fetch specific chunks), but the provided tools are sufficient for the domain.
Read-only verifier for 25 ProofRelay MCP tools and non-confidential evidence bundles.
Verify PyPI and npm packages, symbols, and version diffs against real artifacts. Free, no account.
Read-only FastAPI container assessment with file evidence and an advisory deployment plan.
Verifies Bernstein run receipts and hash chains; lists the shipped presets and adapters. Read-only.
