Merged
Conversation
Remove hardcoded provider map (anthropic/openai) and instead query the running llmspy pod's providers.json for env var names and available providers. This means any provider llmspy supports (zai, deepseek, google, mistral, etc.) works with `obol model setup` without code changes.
Extract pure functions (parseProviderEnvKey, parseAvailableProviders, buildProviderStatus, patchLLMsJSON) from kubectl-calling wrappers so they can be tested without a running cluster. 23 test cases covering parsing, status cross-referencing, JSON patching, and edge cases.
Adds TestIntegration_ZaiInference that exercises a provider NOT in the old hardcoded map, proving zero-code-change provider support. Uses glm-4-flash via llmspy routing with ZHIPU_API_KEY from .env.
Replace the minimal ethereum skill with three production-ready skills: - obol-blockchain: Ethereum JSON-RPC via eRPC with rpc.py helper, ERC-20 reference, contract addresses, ENS, gas estimation - obol-k8s: Kubernetes cluster diagnostics via ServiceAccount API with kube.py helper for pod/service/event introspection - obol-dvt: Obol DVT cluster monitoring with API examples for validator effectiveness, operator auditing, and exit coordination Keeps the hello skill as-is for smoke testing. Test coverage spans every layer of the skills pipeline: - Unit: embed discovery, copy, skip-existing, volume path, staging, injection, no-op without skills - Integration: staging on sync, pod visibility via kubectl exec, skills sync --from, idempotent re-sync, skill-through-inference (agent references loaded skills), Python smoke tests piped into pod
bussyjd
force-pushed
the
oisin/skills
branch
from
0f16d5e to
27855d9
Compare
This was referenced Feb 20, 2026
agaskrobot
approved these changes
Feb 20, 2026
Replace stale ethereum skill references with the four embedded skills (hello, obol-blockchain, obol-k8s, obol-dvt). Add standalone Skills section to README, update CLAUDE.md delivery flow and source file listings, and expand obol-stack-dev skill with skills system coverage.
…nto feat/rich-skills
…tests - Use glm-5 instead of glm-4-flash (not in providers.json) - Pass explicit instance ID to skills sync when multiple instances exist
Change the name on skills and test
Add rich embedded skills with full test coverage
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
3 participants
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.
Refactors how we do default vs petnames in openclaw skills.
Had trouble with the architecture, but think third plans the charm i got it working. Need to look fresh in the morning though.