• Home
  • Features
  • Pricing
  • Docs
  • Announcements
  • Sign In

VolvoxLLC / volvox-bot / 25589077693
91%

Build:
DEFAULT BRANCH: main
Ran 09 May 2026 02:28AM UTC
Jobs 1
Files 324
Run time 1min
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

09 May 2026 02:25AM UTC coverage: 90.391% (-0.03%) from 90.417%
25589077693

push

github

web-flow
feat: triage latency optimization with multi-provider SDK and structured output (#799)

* feat: add multi-provider SDK architecture

Replace hardcoded Anthropic SDK with pluggable multi-SDK registry
supporting minimax, moonshot, and openrouter providers. Restructure
providers.json with sdk/sdkOptionsKey fields and capability flags
(structuredOutput, promptCaching). Add zod for schema validation.

* feat: add structured output schemas and parsing

Extract Zod schemas for classifier and responder output into dedicated
modules (triage-classify-schema.js, triage-respond-schema.js). Add JSON
fallback prompt for models without native structured output support.
Remove inline schema from system prompt for cleaner separation of concerns.

* feat: add triage latency tuning config fields

Centralize numeric constraints in triage-config-fields.js as single
source of truth for bot, API, and dashboard. Add config fields for
debounce, cooldown, memory timeout, prompt caching, context age,
structured output retries, classifier token cap, and direct mention
fast-path. Add model validation to config command to reject invalid
provider:model combinations at save time.

* feat: implement per-user concurrent evaluation and direct mention fast-path

Replace global evaluation lock with per-user inFlight map to enable
concurrent triage evaluations across different users. Route direct
mentions and replies straight to the responder, skipping the classifier
for lower latency. Decouple auto-mod to fire-and-forget so it no longer
blocks the triage pipeline. Optimize context fetching (cap age, reduce
default limit) and convert stats logging to synchronous.

* docs: update documentation for triage latency optimization

Add multi-provider SDK rule to AGENTS.md. Add "Adding an AI Provider"
guide to DEVELOPMENT.md covering SDK setup, providers.json config, and
structured output patterns. Update README triage description, changelog
with breaking default changes (cooldown 10s→0, context ... (continued)

10475 of 12225 branches covered (85.69%)

Branch coverage included in aggregate %.

55 of 60 new or added lines in 11 files covered. (91.67%)

3 existing lines in 1 file now uncovered.

16325 of 17424 relevant lines covered (93.69%)

167.01 hits per line

Uncovered Changes

Lines Coverage ∆ File
2
94.0
-6.0% src/modules/triage-parse.js
1
90.94
-0.49% src/modules/aiAutoMod.js
1
97.62
-2.38% src/modules/triage-config.js
1
93.81
-0.27% web/src/lib/amplitude.ts

Coverage Regressions

Lines Coverage ∆ File
3
83.11
-0.91% src/modules/welcomeDynamicContext.js
Subprojects
ID Flag name Job ID Ran Files Coverage
1 /jobs/181505858 25589077693.1 09 May 2026 02:28AM UTC 324
90.39
GitHub Action Run
Source Files on build 25589077693
  • Tree
  • List 324
  • Changed 11
  • Source Changed 9
  • Coverage Changed 11
Coverage ∆ File Lines Relevant Covered Missed Hits/Line Branch Hits Branch Misses
  • Back to Repo
  • Github Actions Build #25589077693
  • 63517e05 on github
  • Prev Build on main (#25587880327)
  • Next Build on main (#25597669744)
STATUS · Troubleshooting · Open an Issue · Sales · Support · CAREERS · ENTERPRISE · START FREE · SCHEDULE DEMO
ANNOUNCEMENTS · TWITTER · TOS & SLA · Supported CI Services · What's a CI service? · Automated Testing

© 2026 Coveralls, Inc