vergil-project

vergil-project/vergil-claude-plugin

Plugin Claude Code

Plugin marketplace listing 1 plugin: vergil.

vergil-project/vergil-claude-plugin

Plugin Claude Code

Shared hooks, skills, agents, and commands for all managed repositories in the vergil-tooling ecosystem.

vergil-project/vergil-claude-plugin

Hook Claude Code

Runs before the agent uses a tool for Bash tool calls, executing guard.sh. From vergil-project/vergil-claude-plugin.

vergil-project/vergil-claude-plugin

Settings file Claude Code

Agent settings declaring 1 hook event (PreToolUse) and 1 allowed tools.

vergil-project/vergil-claude-plugin

Instructions file CodexOpenCode

AGENTS.md instructions for vergil-project/vergil-claude-plugin, covering standard tooling plugin agent instructions, user overrides (optional), standards reference, shared skills and local overrides.

vergil-project/vergil-claude-plugin

Instructions file

Claude Code instructions for vergil-project/vergil-claude-plugin, covering claude.md, memory management, parallel ai agent development, structure and rules.

vergil-project/vergil-claude-plugin

Agent

Session bootstrap agent. Use proactively at the start of every work session to validate the repository profile, check branch state, and load context. Must run before any code changes.

vergil-project/vergil-claude-plugin

Agent

The plugin provides custom subagents that can be invoked during Claude Code sessions.

vergil-project/vergil-claude-plugin

Hook

Runs after a tool call finishes for Bash tool calls, executing detect-deprecation-warnings.sh. From vergil-project/vergil-claude-plugin.

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Triage a deprecation warning into a tracked decision. Use whenever a deprecation warning surfaces — the detect-deprecation-warnings PostToolUse hook fires, or the human points at a warning in test/CI/build output and asks how to handle it ("triage this deprecation", "what do we do about this warning"). Searches for an…

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Use as the DEFAULT entry point for non-trivial work — building a feature, a cross-cutting or multi-PR initiative, or promoting a brainstorm into the epic/task framework. epic-create is the outer workflow that runs brainstorming, creates the epic and its bookend tasks in the org .github, and drives spec → pushback →…

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Drive a GitHub epic's tasks to their human gates as the USER agent. Use whenever the human wants to start or resume work on an epic as a whole — "implement epic.

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Close out an epic by authoring its terminal retrospective — the backward-looking record that partners the spec and plan. Use when the human wants to finish an epic ("run the retrospective", "close out epic ", "we're done with this epic", "wrap up the epic"). The run-once finishing gate of the three-skill lifecycle…

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Capture or restore work state across sessions. Use when wrapping up and you want to preserve context for next time ("/handoff stop", "save my progress", "hand this off before I stop"), or when picking work back up ("/handoff start", "resume where I left off", "what was I working on"). Captures work state before…

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Run a deployment task end-to-end as the USER agent — a deployment-labelled issue that installs/syncs merged changes so they are usable, then records the result as a comment. Use when the human asks to deploy, install, roll out, or sync merged work ("deploy.

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Implement a GitHub issue end-to-end as the USER agent. Use whenever the human asks to implement, build, fix, or start work on an issue — "implement.

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Run a validation task end-to-end as the USER agent — a validation-labelled issue whose acceptance is a live check, not a code change. Use when the human asks to validate, verify, or run the checklist on a validation issue ("validate.

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Review memory files collaboratively with the human. Use when the human asks to audit, review, clean up, prune, or check memory ("audit my memory", "review MEMORY.md", "are these memories still accurate", "clean up memory"). Verifies each entry, assesses staleness, and routes it to the correct scope with human approval.

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Set up or refresh a project's memory directory. Use when the human asks to initialize, set up, bootstrap, or update memory ("init memory", "set up MEMORY.md", "add the memory policy header", "refresh the memory header"). Creates or updates MEMORY.md with the human-approval policy header.

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Migrate a repo's existing open-issue backlog into the epic/task framework. Use when the human asks to "migrate ", "bring into the framework", "onboard this repo's backlog", "do the migration for ", or wants the existing pile of open issues triaged into epics/tasks/ad-hoc/closed so the roadmap and audit reflect…

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Monitor and reconcile an open PR after vrg-submit-pr. Use whenever a PR has just been opened and needs watching — vrg-submit-pr emits the /vergil:pr-watch line, or the human asks to watch, monitor, or babysit the PR through CI and review. Runs in the USER agent session: block on PR state with vrg-pr-await, reconcile…

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Produce a structured summary in decisions, operations, or stream-of-consciousness (SOC) mode. Use when the human asks for a structured summary, a decision or operations writeup, or SOC capture ("summarize this", "write up these decisions", "capture this as SOC", "log this operation"). SOC mode is the canonical…

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Capture an uncurated idea, bug, or to-do into the intake queue so it is never lost. Use whenever the human says "don't forget X", "capture this", "note a todo", "file an issue for…", or riffs an idea they want to come back to later — especially mid-task or via voice. Creates an intake issue — triage (a problem not yet…

vergil-project/vergil-claude-plugin

Skill Claude CodeCodex

Groom the triage intake queue — review uncurated issues and route each into the epic/task model. Use when the human asks to "review triage", "groom the backlog", "process the triage queue", "what's in triage", or wants the periodic (roughly weekly) intake pass. Collects every triage-labelled issue across the org and…