graphhelm
claudecodexUnderstand GraphHelm and choose two evidence-backed next actions.
6 skills · 1 MCP server · hooks
Keel, code contracts, evidence retrieval, and advisory memory skills.
Plugin graphhelm-development-contracts from marketplace graphhelm in stabem/GraphHelm. Commands read from the repo's manifest — today.
blind-reviewReview one PR blind, as the single assigned reviewer: read the earlier comments, run the tests the change reaches and the Keel check, try to break the user-visible promise, and post APPROVE, APPROVE-WITH-RISK or BLOCK with the identity line. Use when you are assigned as a PR's reviewer.
implementPlan and write a bounded change under Keel in one step: decide the paths, promise and proof kind, write the card, stay inside the card's paths, gather only the context the promise needs and cite it, add only what the promise needs, and keep a failing-first test only for invariants. Use for a bounded
keelUse when deciding which files a change may touch (a bug found in a neighbouring file, an unrelated cleanup, anything outside the card's paths), when asked whether a promise can be reported as verified without a real observer on this host, or when deciding whether a new test proves a fix (for example
memory-curatorPropose durable lessons as advisory candidates after work lands, each tied to the state in which it was true. Use when a task produced a defect class, a measurement, or a correction worth keeping, and never to record it directly.
mergeMerge an approved PR at the exact head the review read, check its closing keywords, read back what landed, and release the workspace. Use only as the PR's approving reviewer, right after the approval.
task-planFirst step of a task after its issue: decide the paths, promise, proof kind (journey, tests, both, none), review count and skills for the change, and record them as one keel.plan record. Use before code when the change touches a journey screen, a new public surface or a persistence/permissions/compa
test-auditUse when someone asks whether a test proves anything, wants existing tests judged or audited, wants to delete a slow or flaky test, or is about to add or change a test: four questions gate each new test, an audit lists the tests that prove nothing (self-comparison, a mock confirming itself), and eac
Understand GraphHelm and choose two evidence-backed next actions.
6 skills · 1 MCP server · hooks
Keel's plan, execute, review and merge skills for any project; no Runtime, MCP or journeys.
6 skills
Journey-Proven Development skills.
3 skills
Codex session hook compatibility companion for GraphHelm versions that skip hooks in the Agent Plugin manifest.
1 MCP server
Draws a GraphHelm journey run inline in the Claude Code chat, with Approve and Reject.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
15 skills · hooks
Scaffold, validate, and publish new Claude Code plugins with proper structure
3 skills · 1 command · 1 agent
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
Use Alibaba Open Code Review from local Codex.
2 skills
Claude Code skill for creating and presenting Slidev slidedecks
1 skill
Code intelligence powered by a knowledge graph. Provides execution flow tracing, blast radius analysis, and augmented search across your codebase.
12 skills · 1 MCP server · hooks
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.