Repository navigation
Make defaults match across the CLI, TUI, desktop app and MCP server - #503
Merged
Merged
Conversation
Contributor
|
Site preview: https://pr-503.netscli-site-preview.pages.dev Built from e6e7b2b with Production is unaffected: netscli.com is served from GitHub Pages via |
- Desktop scan and inspect default to the core's 22,80,443 instead of a five-port list of their own; sweep reuses the same constant. - MCP: one ops_config() helper. Without a client timeout each phase keeps the core default (ping 1000, scan 500, DNS 1500) instead of the tool's single advertised default standing in for all three. - MCP schemas stop advertising defaults the code does not apply (dns type A, subnet 192.168.1.0/24, start_pcap_capture outputFile). - Ping count capped at MAX_PING_COUNT in the TUI (was unbounded), the Tauri command (was 64) and the desktop form (was 50). The MCP mDNS clamp uses MAX_MDNS_TIMEOUT_MS instead of a literal. - Desktop inspect passes the concurrency preference through. - Desktop pcap duration defaults to 10 s like the core. Left alone on purpose: MCP ping count defaults to 1 (single summary for a model, documented in schemas.rs), the desktop keeps its 1,000-packet cap (its results table is not virtualised), and the desktop resolves hostnames on discover and sweep (a UI choice; the CLI keeps it opt-in for speed).
fstubner
force-pushed
the
fix/consistent-defaults
branch
from
October 3, 2026 19:41
1fe3c04 to
55f86c5
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Found by reading every default for every operation across the four interfaces and the core constants.
Changed
22,80,443,8080,8443; everything else scans the core's22,80,443. Now the same.ops_config()helper: a clienttimeoutstill applies to every step, and without one each step keeps the core default. Two tests pin both cases.A(applies all types), subnet192.168.1.0/24(applies the local subnet),start_pcap_captureoutputFile: capture.pcap(applies a per-job name). Replaced with descriptions of what happens.MAX_PING_COUNT(256).Ops::default()); now passes it like scan/discover/sweep.Left alone on purpose
Checks:
cargo clippy -D warningsand tests for netscli-mcp, netscli and netscli-gui pass locally. The desktop JS lint and unit tests did not run locally (incomplete node_modules); CI runs them. Two desktop tests that pinned the five-port list are updated.