chunk-invariance
A streaming filter (a redactor, a guardrail, a rewriter) is correct only if, for every way of
splitting an input into chunks, the concatenated streamed output equals the output of filtering
the whole input at once. This package turns that rule into one assertion.
from chunk_invariance import assert_chunk_invariant
def redact(text):
return text.replace("sk-AbCd1234", "[KEY]")
def redact_stream(chunks):
for chunk in chunks:
yield redact(chunk)
def test_streaming_redaction():
assert_chunk_invariant(redact_stream, redact, "use sk-AbCd1234 now")
That test fails, and names the smallest split that breaks it:
chunk_invariance.core.ChunkInvarianceError: stream filter is not chunk-invariant
input: 'use sk-AbCd1234 now'
split: ['use s', 'k-AbCd1234 now'] (cut at 5)
whole: 'use [KEY] now'
streamed: 'use sk-AbCd1234 now'
split #6 in the order tried; generated splits run fewest parts first
pip install chunk-invariance
pip install "chunk-invariance[hypothesis]"
pip install "chunk-invariance[mcp]"
What it catches
- A filter that evaluates each chunk alone, so a value cut across two chunks matches neither half.
- A filter that prepends a fixed carryover from the previous chunk. The match fires, but the start
of the value was already emitted with the previous chunk.
- A hold-back that is shorter than the pattern, a missing end-of-stream flush, and output that is
corrupted rather than leaked (any difference from the whole-input output fails).
chunk_invariance.examples has one correct filter (hold_back) and the two wrong designs
(design_a, design_b) for the same pattern, as a worked comparison.
What it does not catch
It tests a filter you can call in-process. It does not test a running gateway, where the chunks
are decided by an upstream model, a network and an SSE parser. For that, use
pii-leak-benchmark, which sends
fragmented responses through the gateway and checks what reaches the client.
It checks the splits you ask for. The default is the whole input plus every single cut, which
catches a value cut in two. splits="all" tries every split of a short input (up to 16
characters); splits=3 tries every split into at most three parts; an explicit list of chunk
lists runs exactly those. It cannot prove a filter correct for inputs you did not give it.
Use it from your AI assistant
Do you use Claude Code, Cursor or Claude Desktop? The package has an optional MCP server. With it
you can ask the assistant to run the every-split check on your own streaming function, and then
to add the test to your repo.
pip install "chunk-invariance[mcp]"
claude mcp add chunk-invariance -- chunk-invariance-mcp
For Claude Desktop, add this to claude_desktop_config.json (give the full path to the command
if Claude Desktop cannot find it on your PATH):
{
"mcpServers": {
"chunk-invariance": { "command": "chunk-invariance-mcp" }
}
}
Then ask, for example: "Check whether redact_stream in myapp/filters.py leaks values split
across chunks, and add a regression test for it."
There are three tools. check_stream_filter imports your function (myapp.filters:redact_stream),
streams each sample through it in every two-part split and tells you which split broke it, for
example that 'sk-AbCd1234' came out unredacted when the input was cut into 'use s' and
'k-AbCd1234 now'. write_regression_test returns a pytest file built on assert_chunk_invariant,
so the check stays in your CI. explain is one paragraph on why per-chunk filtering leaks.
The default samples (an email, an SSN, a card number and a few API-key shapes) are all fake. Pass
your own made-up samples in the formats your filter redacts, such as a key you typed yourself in
your provider's format. A sample the filter does not change tests nothing, and the tool says so.
Never pass a real secret or real personal data as a sample. The samples, and any value your filter
leaks, come back in the tool's result, and your assistant sends that result to its model provider
like the rest of the conversation. So does the message of any exception your code raises, but
nothing your code prints. The same samples also end up in the generated test file.
Checking a function runs your code. The server imports your module on your own machine, in a
separate process that is stopped after a timeout (60 seconds by default). If your project has a
.venv (or venv), that is the Python it uses; the python argument accepts only that
interpreter or the server's own. The target, flush_target and whole_target must be Python
functions defined in source files under project_dir: built-ins, the standard library and
installed packages (anything in the interpreter's site-packages, including your .venv's) are
refused, and so is a project_dir that is or contains the interpreter's own library. The check
process gets a minimal environment (PATH, HOME, temp and locale variables), not the server's,
so a filter that reads its configuration from other environment variables will not see them.
project_dir is chosen by the assistant's tool call, so whatever code lives under it is what can
run: approve calls only for your own project. The server itself makes no network calls, sends no
telemetry and writes no files, though the code it runs can do whatever that code does. Saving the
test file is up to you and your assistant.
API
| Name | Does |
|---|
assert_chunk_invariant(stream_filter, whole_filter, text, *, splits="all-two-part") | Raises ChunkInvarianceError (an AssertionError) on the first failing split, fewest parts first. Returns the number of splits checked. Works on str and bytes. |
assert_chunk_invariant_async(...) | The same, for an async generator filter, inside an async test. |
two_part_splits(text) | Every split into two non-empty parts. |
all_splits(text, max_parts=None) | Every split into at most max_parts non-empty parts, the whole input first. |
per_chunk(process, *, state=dict, flush=None) | Adapts process(chunk, state) -> output and flush(state) -> output, with fresh state per stream. |
from_async(async_stream_filter) | Adapts an async generator filter for the synchronous assertion. |
strategies.splits_of(text, max_parts=None) | Hypothesis strategy: random splits of text, shrinking toward fewer cuts. |
strategies.text_and_splits(texts, max_parts=None) | Hypothesis strategy: (text, chunks) pairs. texts is itself a strategy, such as st.sampled_from([...]), not a list. |
stream_filter is called once per split, so it must start from fresh state each time.
Background
The invariant and the two wrong designs are described in Phalak, N. (2026), Split-Boundary
Leaks in Streaming Guardrails, Zenodo, 10.5281/zenodo.22909585.
Standard library only (the mcp extra adds the MCP SDK, for the server alone). Apache-2.0. A TypeScript version with the same API is published on npm
as chunk-invariance.