Commit efbecd4
release: v0.3.1 — string pattern + max_length, empirical study, latency A/B
User-facing additions in v0.3.1:
Param(
...
pattern: str | None = None, # re.fullmatch
max_length: int | None = None, # char count
)
Both optional, backwards-compatible; enforced at the Bridge.
Caught by an empirical study that fed 295 real LLM-emitted tool calls
(DeepSeek V3 + Qwen 2.5-72B via SiliconFlow) through every protocol's
schema-layer validator. The same study now appears as figure 4 in the
paper and as the prose in section 3; the v0.3.1 pattern + max_length
fields close the prompt-injection gap (DCP 30% -> 50%, matching
OpenAPI's expressiveness at three orders of magnitude less footprint).
Also adds tools/bench_latency_iotmcp.py and an iotmcp_echo Arduino
sketch: same ESP32-S3, same UART, same asyncio host stack, with the
device side speaking newline-delimited JSON like IoT-MCP's reference
MCU servers. DCP 15.60 ms vs IoT-MCP 15.59 ms median over 1000
round-trips — capability scoping and full schema validation are free
at this scale.
Bug fix also worth a release: examples/smart_panel_manifest.yaml's
play_tone duration param was type "duration" (which the safety
layer coerces to float, breaking the firmware's int reader). Now
type "int" with unit "ms".
Includes the rebuilt arXiv tarball (docs/paper/dcp-arxiv-v0.3.1.tar.gz)
that goes with this release. See CHANGELOG for the full list.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>1 parent 8f8005c commit efbecd4
4 files changed
Lines changed: 50 additions & 2 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
8 | 8 | | |
9 | 9 | | |
10 | 10 | | |
| 11 | + | |
| 12 | + | |
| 13 | + | |
| 14 | + | |
| 15 | + | |
| 16 | + | |
| 17 | + | |
| 18 | + | |
| 19 | + | |
| 20 | + | |
| 21 | + | |
| 22 | + | |
| 23 | + | |
| 24 | + | |
| 25 | + | |
| 26 | + | |
| 27 | + | |
| 28 | + | |
| 29 | + | |
| 30 | + | |
| 31 | + | |
| 32 | + | |
| 33 | + | |
| 34 | + | |
| 35 | + | |
| 36 | + | |
| 37 | + | |
| 38 | + | |
| 39 | + | |
| 40 | + | |
| 41 | + | |
| 42 | + | |
| 43 | + | |
| 44 | + | |
| 45 | + | |
| 46 | + | |
| 47 | + | |
| 48 | + | |
| 49 | + | |
| 50 | + | |
| 51 | + | |
| 52 | + | |
| 53 | + | |
| 54 | + | |
| 55 | + | |
| 56 | + | |
| 57 | + | |
| 58 | + | |
11 | 59 | | |
12 | 60 | | |
13 | 61 | | |
| |||
Binary file not shown.
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1 | 1 | | |
2 | 2 | | |
3 | | - | |
| 3 | + | |
4 | 4 | | |
5 | 5 | | |
6 | 6 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
8 | 8 | | |
9 | 9 | | |
10 | 10 | | |
11 | | - | |
| 11 | + | |
12 | 12 | | |
13 | 13 | | |
14 | 14 | | |
| |||
0 commit comments