little-canary
Blocks Gemini CLI agent runs when a local Little Canary server rejects the prompt.
Open source Open in the app JSON README (API)
About
Blocks Gemini CLI agent runs when a local Little Canary server rejects the prompt.
Details
- Kind
- Plugins
- Topic
- AI, RAG & memory
- Publisher
- hermes-labs-ai
- Origin
- gemini
- Category
- ferramentas
- Version
- 0.3.5
- Stars
- 28
- Forks
- 4
- Last push
- 2026-09-07T06:56:04Z
- Repository state
- ativo
- Language
- Python
- License
- Apache-2.0
- Added
- 2026-09-03 09:00:31
- Updated
- 2026-09-03 09:00:31
- Origin id
hermes-labs-ai/little-canary
README
# little-canary
Prompt-injection sensing through a powerless sacrificial model.
**Links:** [Website](https://littlecanary.ai) · [Hermes Labs product page](https://hermes-labs.ai/little-canary)
Little Canary lets untrusted language affect a small model with no application tools or authority, then inspects that model's response for compromise residue before your agent acts. Structural checks catch known input shapes; the distinctive behavioral layer asks what the input *did to the canary*.
```text
untrusted text
→ structural preflight
→ powerless sacrificial model
→ response-residue analysis
→ route: PASS / FLAG / BLOCK, with explicit coverage state
```
Little Canary is an inbound risk sensor, not a security guarantee or an agent runtime.
## Technical note
[Behavioral Canarying for Prompt Injection: Powerless Model Probes with Explicit Coverage Semantics](https://hermes-labs.ai/research/behavioral-canarying)
documents Little Canary's pre-execution sensing architecture and the separation
between routing disposition and inspection coverage. It does not claim
universal detection, formal security, or aggregate accuracy for the current
release. Cite the version-independent concept DOI at
[10.5281/zenodo.21818564](https://doi.org/10.5281/zenodo.21818564):
```bibtex
@misc{bosch2026behavioralcanarying,
author = {Bosch, Rolando},
title = {Behavioral Canarying for Prompt Injection: Powerless Model
Probes with Explicit Coverage Semantics},
year = {2026},
publisher = {Zenodo},
doi = {10.5281/zenodo.21818564},
url = {https://doi.org/10.5281/zenodo.21818564},
note = {Technical note}
}
```
See [hermes-publications/papers/behavioral-canarying](https://github.com/hermes-labs-ai/hermes-publications/tree/main/papers/behavioral-canarying)
for the full evidence boundary.
## Release truth
Source checkouts, GitHub releases, and registry builds are separate evidence
surfaces. The version of the source you are reading is recorded in this
repository's own metadata (`pyproject.toml` and
`little_canary/__init__.py`); this README does not assert what any registry
holds at the moment you read it. For current publication state, consult the
live authorities:
[GitHub Releases](https://github.com/hermes-labs-ai/little-canary/releases)
and [PyPI](https://pypi.org/project/little-canary/). Historically, GitHub
`v0.3.1` was source-only and `0.3.2` was intentionally not published or
reused. The `demo` commands documented below require `0.3.3` or later; verify
the installed artifact with `little-canary --version` and confirm it matches
the version you intended to install.
## Install
This source tree supports Python 3.9–3.13. A published artifact's own
package metadata is the authority for the Python range that artifact
advertises.
From the registry (see PyPI for available versions):
```bash
python -m pip install little-canary
little-canary --version
```
For development from a source checkout:
```bash
python -m pip install .
little-canary --version
```
## Run the evidence gates without writing Python
### Replay gate: zero egress
This release, like `0.3.3` before it, deliberately packages no replay
fixture. The
available historical live transcript is incomplete, so turning it into a
fixture would fabricate missing provenance and response bytes. Therefore this
exact build reports `REPLAY UNAVAILABLE` and exits `2`:
```bash
little-canary demo --replay
```
That is a release hold, not a clean verdict. It makes no model or network call,
does not report risk `0`, and does not silently fall back to live mode.
After a complete dedicated live capture is admitted and packaged, the same
command will re-run the shipped analyzer over its versioned clean/attack
response pair. Its first lines will state:
```text
RUN_KIND REPLAY
MODEL_CALL no — recorded output
CANARY NOT EXERCISED THIS RUN
EGRESS none
```
Success is `REPLAY VERIFIED`: the recorded capture exercised a canary, the current command did not, and the analyzer reproduced the expected contrast. Replay does not prove that a model is installed, reachable, or currently behaves the same way.
A build without an admitted complete capture exits `2` with `REPLAY UNAVAILABLE`; it never invents response bytes or makes a hidden live call.
### Live proof gate: explicit local egress
Live mode requires an endpoint dedicated to this evaluation. A shared or
unleased runtime is not release evidence; leave the gate unevaluated instead
of commandeering it.
```bash
little-canary demo --live \
--backend ollama \
--model qwen2.5:1.5b \
--endpoint http://127.0.0.1:11434
```
Live mode uses a fixed synthetic clean/attack pair and disables the structural filter so the demonstration tests the behavioral mechanism. Before sending either prompt it prints the backend, model, redacted loopback origin, and that raw synthetic input will leave the process. It does not accept arbitrary input and does not fall back to replay.
Results:
- exit `0`: complete clean/non-block plus attack/block contrast;
- exit `1`: complete calls but `NO CONTRAST` or analyzer expectation mismatch;
- exit `2`: invalid usage, unavailable model/backend, protocol failure, or otherwise incomplete/degraded run.
Add `--json` for the agent-readable result. Bare `little-canary demo` exits `2` and requires an explicit `--replay` or `--live` choice.
## Python API
```python
from little_canary import SecurityPipeline
pipeline = SecurityPipeline(
canary_model="qwen2.5:1.5b",
mode="full",
)
verdict = pipeline.check(untrusted_text)
if verdict.degraded:
# Fail-open routing may still be safe=True, but behavioral coverage failed.
quarantine_or_apply_your_availability_policy(untrusted_text)
elif not verdict.safe:
block(untrusted_text, verdict.summary)
else:
forward_to_agent(verdict.safe_input)
```
Routing and evidence are separate:
| Field | Meaning |
|---|---|
| `safe` | Whether configured routing policy allows forwarding |
| `degraded` | Whether an enabled required inspection dependency failed |
| `canary_status` | `exercised`, `failed`, `disabled`, or `skipped_after_block` |
| `analysis_method` | `regex`, `llm_judge`, or `none` |
| `analysis_status` | `exercised`, `failed`, or `not_applicable` |
| `canary_risk_score` | Measured risk, or `None` when no valid measurement exists |
Fail-open is availability-first, not a clean verdict. If an enabled canary fails, Little Canary may return `safe=True`, but it also returns `degraded=True`, `canary_status="failed"`, risk `None`, and no PASS label. A failed or skipped layer is never serialized as `passed=true`.
Callbacks follow the same truth boundary: `on_degraded` and `on_unexercised`
are distinct from `on_pass`. `CanaryGuard` and audit records propagate
degraded, `STRUCTURAL_ONLY`, and `UNSCREENED` state.
## Backends and data flow
The library supports local Ollama and OpenAI-compatible endpoints. The demo intentionally supports loopback Ollama only.
- The canary backend receives the raw input and the known canary system prompt.
- If an optional LLM judge is configured, it receives the raw input and canary response.
- A remote endpoint therefore sends data off-machine.
- AuditLogger omits raw input but stores an unsalted SHA-256 input hash. That supports correlation; it is not anonymity.
- Runtime inspection found no separate product telemetry path, but provider requests are still egress.
HTTP `200` alone is not successful model coverage. Missing, empty, null, non-string, malformed, timeout, and transport responses are visible protocol failures. Provider bodies, credentials, URL userinfo, and query strings are not included in public errors.
## What “powerless” means
Little Canary does not give the canary model application tools, credentials, or
output execution. The default `SecurityPipeline` strips response bytes and
signal-evidence excerpts from its layer snapshot before callbacks or JSON
serialization; it does not automatically forward canary output to an
authoritative agent.
The low-level `CanaryProbe` and `AnalysisResult` APIs return or retain the
response because analysis requires it. Treat those objects as sensitive: do
not execute or forward their contents, and do not attach authority-bearing
tools to the canary runtime.
This is a library-level capability boundary, not an operating-system sandbox. If your deployment wraps the model with tools or forwards its output elsewhere, that deployment changes the claim.
## Local HTTP adapter
```bash
little-canary serve \
--port 18421 \
--mode advisory \
--canary-model qwen2.5:1.5b \
--ollama-url http://127.0.0.1:11434
```
The server binds to `127.0.0.1`, exposes `GET /health` and `POST /check`, and is unauthenticated. Treat it as a local adapter, not a production gateway.
```bash
curl -sS http://127.0.0.1:18421/check \
-H 'Content-Type: application/json' \
-d '{"text":"untrusted text"}'
```
Every accepted non-empty string reaches the pipeline, including one-character input. Malformed, missing, wrong-type, empty, and oversized requests are explicit errors. Text is never silently truncated before inspection. `/health` is liveness-compatible HTTP `200` and includes truthful `ready`, `degraded`, backend, model, and coverage details.
The loopback server has no authentication, TLS, concurrency hardening, or remote-deployment design in this release.
## Gemini CLI extension
This repository is also a Gemini CLI extension. It uses Gemini CLI's
`BeforeAgent` hook to screen the exact current prompt before the agent loop and
deny the run when Little Canary returns `safe: false`.
Start the loopback server in blocking mode, then validate and install a source
checkout. This integration was verified with Gemini CLI 0.32.1:
```bash
little-canary serve --mode block
gemini extensions validate .
gemini extensions install . --consent
```
The hook calls only `http://127.0.0.1:18421/check` by default and completes its
request within three seconds. Transport errors, malformed responses, and
unexercised behavioral coverage are visibly fail-open by default; they are not
reported as a clean pass. Set `LITTLE_CANARY_FAILURE_MODE=deny` in the Gemini
process environment for fail-closed behavior. `LITTLE_CANARY_ENDPOINT` may
select another loopback HTTP `/check` URL, and `LITTLE_CANARY_TIMEOUT_MS` may be
set from 100 through 5000.
This extension blocks one Gemini agent run at its pre-agent boundary. It does
not establish a general security guarantee or replace least privilege and tool
policy.
## Claude Code plugin
This repository is also a Claude Code marketplace that serves one plugin. The
plugin uses Claude Code's `UserPromptSubmit` hook to screen the exact submitted
prompt before the turn starts and block it when Little Canary returns
`safe: false`.
The marketplace manifest lives at `.claude-plugin/marketplace.json`, where
Claude Code discovers it. The plugin itself is the self-contained directory
`plugins/claude-code/` (plugin manifest, hook registration, and the standalone
adapter script). Claude Code copies only that directory into its plugin cache,
so the Gemini CLI extension files at the repository root (`gemini-extension.json`,
`hooks/hooks.json`) are never installed or loaded by Claude Code.
Start the loopback server in blocking mode, then add this repository as a
marketplace and install the plugin. This integration was verified with Claude
Code 2.1.261:
```bash
little-canary serve --mode block
claude plugin marketplace add hermes-labs-ai/little-canary
claude plugin install little-canary@hermes-labs
```
To validate a source checkout, validate both manifests explicitly. Running
`claude plugin validate .` from the repository root only validates the
marketplace manifest, because the repository root is not itself a plugin:
```bash
claude plugin validate .claude-plugin/marketplace.json --strict
claude plugin validate plugins/claude-code --strict
```
To install from a local checkout instead of GitHub, pass the checkout path to
`claude plugin marketplace add` and then run the same install command.
The hook calls only `http://127.0.0.1:18421/check` by default and completes its
request within three seconds. It sends the prompt as the JSON body
`{"text": ...}`. The loopback server rejects request bodies larger than 64 KiB
(65,536 bytes) with HTTP `413`; that ceiling applies to the encoded JSON body,
not to the prompt's character count, so non-ASCII text reaches it sooner. Such
prompts are not screened at all. Transport errors, HTTP errors including that
`413`, malformed responses, and unexercised behavioral coverage are visibly
fail-open by default: the turn continues and Claude Code shows a warning such
as `Little Canary screening unavailable: HTTPError`, so they are never reported
as a clean pass. Set `LITTLE_CANARY_FAILURE_MODE=deny` in the Claude Code
process environment to block the turn on every one of those failures instead.
`LITTLE_CANARY_ENDPOINT` may select another loopback HTTP `/check` URL, and
`LITTLE_CANARY_TIMEOUT_MS` may be set from 100 through 5000. The hook never
writes model context; it emits exactly one JSON object per event.
This plugin blocks one Claude Code turn at its prompt-submission boundary. It
does not screen tool results, and it does not establish a general security
guarantee or replace least privilege and tool policy.
## OpenAI Agents SDK input guardrail
The optional `little_canary.openai_agents` module wraps a `SecurityPipeline`
(or any object with `check(text) -> PipelineVerdict`) as a native Agents SDK
`InputGuardrail`. It runs before the agent starts by default.
```bash
python -m pip install "little-canary[openai-agents]" # Python 3.10+
```
```python
from agents import Agent
from little_canary import SecurityPipeline
from little_canary.openai_agents import little_canary_input_guardrail
pipeline = SecurityPipeline(canary_model="qwen2.5:1.5b", mode="block")
agent = Agent(name="assistant", input_guardrails=[little_canary_input_guardrail(pipeline)])
```
`output_info` carries a `coverage` label: `unsafe` trips the SDK tripwire;
`safe` means behavioral coverage was exercised and clean; `flagged`,
`degraded`, and `unexercised` are visibly not a PASS. Degraded and unexercised
coverage is fail-open by default, matching the pipeline. Pass
`on_degraded="fail_closed"` to trip the wire unless coverage is exercised
`safe`. See `examples/openai_agents_example.py`. Offline tests were exercised
against `openai-agents` 0.22.0; importing `little_canary` never requires it.
The fail-closed option is an explicit caller policy at the SDK boundary. It
does not change `SecurityPipeline` routing or its default fail-open behavior;
it blocks flagged, degraded, and unexercised outcomes as documented above.
Only user-message text is screened. Tool outputs and non-text content are
outside this adapter's coverage, and SDK input guardrails run only for the
first agent in a chain.
## Evidence labels and limitations
Behavioral evidence is labeled:
- `LIVE`: a model call observed for one exact runtime/model/configuration;
- `REPLAY`: analyzer behavior over recorded bytes;
- `MOCK`: controlled protocol or state logic;
- `STATIC_ONLY`: source/artifact inspection without a model call.
These labels are not interchangeable. Temperature zero and a seed can improve repeatability but do not guarantee identical model output or classifications across versions, runtimes, or hardware.
This README makes no aggregate detection, false-positive, latency, or
token-savings claim. Historical benchmark artifacts remain under `benchmarks/`
with their limitations and are not a performance certificate for this
release.
Little Canary should be combined with least privilege, tool policy, data boundaries, monitoring, and output/runtime controls. It does not prove an input harmless, prevent every injection, or replace containment.
## Development
```bash
pytest
ruff check little_canary tests
mypy little_canary # diagnostic until the recorded baseline debt is resolved
python -m build
python -m twine check dist/*
```
Tests are offline by default and mock network behavior. Live evaluation must use a dedicated endpoint that is not serving another workload.
See [SECURITY.md](SECURITY.md) for vulnerability reporting and [benchmarks/README.md](benchmarks/README.md) for the current evaluation boundary.
## License
Apache-2.0. See [LICENSE](LICENSE).