Development
Skill
/opcore
Use for repository code intelligence and introduced-change validation in Zeroshot.
Install
$ npx -y skills add the-open-engine/zeroshot --skill opcore --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
- You can call itInvoke it directly when you want it.
- Slash command
/opcore
Context preview
The summary Claude sees to decide when to auto-load this skill.
Use for repository code intelligence and introduced-change validation in Zeroshot.
SKILL.md
opcore.SKILL.mdname: opcore
description: Use for repository code intelligence and introduced-change validation in Zeroshot.
Opcore
Opcore is Zeroshot's packaged, repository-local constraints gate. It supplements rather than replaces the repository's lint, type, test, CI, and review guardrails.
- Run `npm run opcore:status` before broad work. If the persistent graph is stale, run `npm run opcore:graph:build`; after source changes, refresh it with `npm run opcore:graph:update`.
- Use `opcore graph search`, `impact`, and `review-context`, plus `opcore inspect definition`, `references`, `signature`, and `implementations`, when graph-backed evidence will reduce uncertainty.
- Use `opcore validate hypothetical` or the installed pre-write hook before a proposed write when the harness does not invoke the hook automatically.
- Run `npm run opcore:check` before finalizing source edits. This gate reports only diagnostics introduced relative to the Git base; existing repository debt is not a blocking result.
- Run `npm run opcore:scan` and `npm run opcore:measure` for named repository metrics and deltas, not as opaque ratings.
- Treat unsupported stacks, unavailable tools, stale graphs, and indeterminate results honestly; never report them as clean coverage.
- Preserve all existing repository guardrails. A full-tree `opcore check all` is an audit command, not a blocking commit, CI, or agent gate.
Read more
name: opcore description: Use for repository code intelligence and introduced-change validation in Zeroshot.
Opcore
Opcore is Zeroshot's packaged, repository-local constraints gate. It supplements rather than replaces the repository's lint, type, test, CI, and review guardrails.
- Run `npm run opcore:status` before broad work. If the persistent graph is stale, run `npm run opcore:graph:build`; after source changes, refresh it with `npm run opcore:graph:update`.
- Use `opcore graph search`, `impact`, and `review-context`, plus `opcore inspect definition`, `references`, `signature`, and `implementations`, when graph-backed evidence will reduce uncertainty.
- Use `opcore validate hypothetical` or the installed pre-write hook before a proposed write when the harness does not invoke the hook automatically.
- Run `npm run opcore:check` before finalizing source edits. This gate reports only diagnostics introduced relative to the Git base; existing repository debt is not a blocking result.
- Run `npm run opcore:scan` and `npm run opcore:measure` for named repository metrics and deltas, not as opaque ratings.
- Treat unsupported stacks, unavailable tools, stale graphs, and indeterminate results honestly; never report them as clean coverage.
- Preserve all existing repository guardrails. A full-tree `opcore check all` is an audit command, not a blocking commit, CI, or agent gate.
Ships withzeroshot
Independent executor–verifier orchestration for software changes.
Get the whole plugin
Stats
1,697
Stars
150
Forks
Active
Maintenance
JavaScript
Language
MIT
License
15h ago
Last commit
7mo ago
Created
Repo: the-open-engine/zeroshot