vrr/openclaw
2
0
Fork 0
mirror of https://github.com/openclaw/openclaw.git synced 2026-10-03 09:39:25 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 30
openclaw/extensions/typesafe
History
Peter Steinberger ede3447b10
chore(deps): refresh dependencies with seven-day cutoff (#158298) 
* chore(deps): refresh dependencies with seven-day cutoff

* fix(deps): complete dependency integration checks

* chore: tighten Discord assertion allowances

* test(acpx): align Claude adapter version references

* chore(deps): align remaining dependency references

* fix(deps): preserve schema diagnostics with escaped pointers

* fix(deps): decode pointer paths across schema diagnostics

* build(deps): carry refresh through main changes

* test(bench): isolate healthy probe from scheduler delay
2026-09-26 20:42:53 -07:00
..
assets improve(plugins): give bundled logos consistent white icon tiles (#155259) 2026-09-23 19:09:26 -07:00
src improve(typesafe): defer startup work until the first decision (#157380) 2026-09-25 07:23:01 +08:00
index.ts feat(decisions): add explicit evaluation and Labs opt-in foundation (#155134) 2026-09-22 15:55:02 -07:00
LICENSE
openclaw.plugin.json feat(decisions): add explicit evaluation and Labs opt-in foundation (#155134) 2026-09-22 15:55:02 -07:00
package.json chore(deps): refresh dependencies with seven-day cutoff (#158298) 2026-09-26 20:42:53 -07:00
README.md feat(decisions): add explicit evaluation and Labs opt-in foundation (#155134) 2026-09-22 15:55:02 -07:00

README.md

TypeSafe AI for OpenClaw

Official external plugin for typed decisions with hosted TypeSafe AI Jev models or a local Kev System One server. It provides Choice, Score, and Boolean judgments through OpenClaw's shared decision-model API. Core supplies the provider-neutral decision_evaluate agent tool when the agent has an effective decisionModel selection, subject to normal tool policy and harness capabilities. No separate tool enablement is required.

Requires OpenClaw and plugin API 2026.9.6 or later. Released OpenClaw 2026.9.5 does not include the decision API.

openclaw plugins install @openclaw/typesafe

The ClawHub install spec is clawhub:@openclaw/typesafe. First publication is pending a supporting release. Enable the plugin, configure a protected TypeSafe credential, and select typesafe/jev-latest as your agent's decisionModel. With that selection, decision_evaluate sends supplied evidence to TypeSafe AI and incurs its normal usage charges. Other providers route evidence according to their own configuration.

For local Kev, configure plugins.entries.typesafe.config.baseUrl with the server's loopback origin, such as http://127.0.0.1:8009, omit apiKey, and select typesafe/kev-latest. The plugin calls /v1/systemone without a hosted credential. Start the server separately with your chosen checkpoint; the decision-model selection labels requests and does not download or load a model.

See the TypeSafe AI setup guide and decision-model documentation for configuration, rubrics, and API semantics.