claude

apex-dispatch

Route and govern delegation: a compiled routing policy picks the tier, provider, fan-out and review shape for each task or ad-hoc ask; PreToolUse/SubagentStop hooks enforce it; provider shims run external workers; every route, spawn, worker run and verdict lands in a hash-chained ledger, which report.sh --per-task prices per routed task. Drives apex-scope-loop as its execution engine.

1 on GitHubby Apex Insightsv0.7.0MITupdated 16 days agodevelopment
/

Plugin apex-dispatch from marketplace apex-dev-skills in Skobyn/Apex-Dev-Skills. Commands read from the repo's manifest — 16 days ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (2)

  • dispatch-route

    Read an apex-dispatch ROUTE block (from route.sh, or between TAGS and STATUS in an apex-scope-loop iterate brief) and dispatch exactly what it says — only roster roles, as the generated apex-dispatch agents, at the routed model and effort, within the routed fan-out, lanes and budgets, with the revie

  • dispatch-worker

    Brief an external or separate-session worker (claude -p, codex) for an apex-dispatch route whose ROUTE_PROVIDER is not claude-session, run it through its shim (${CLAUDE_PLUGIN_ROOT}/bin/worker-codex.sh or ${CLAUDE_PLUGIN_ROOT}/bin/worker-claude-p.sh), record a reviewer's verdict with checkpoint.sh r

Slash commands (6)

  • /apex-dispatch:compile

    Compile apex-dispatch's routing policy (compile.sh) into its committed artifacts — compiled policy, generated agents, settings snippet, hooks.json — or check them for staleness (--check), print this repo's overlay-merged policy (--print-merged), validate an overlay (--overlay PATH), or render the co

  • /apex-dispatch:doctor

    Run apex-dispatch's preflight (doctor.sh) — claude version, the sibling apex-scope-loop, apex-guardrails, compile --check, the merged policy, provider binaries, CLAUDE_CODE_SUBAGENT_MODEL, the settings snippet, versions — and write doctor.json, which routing reads to choose degrade paths. Pass an op

  • /apex-dispatch:done

    Close the active apex-dispatch task — verify the ledger chain (and, under enforcement, the route evidence), then checkpoint.sh complete for a plan task (gate, review, tier and G12 all bound to HEAD), or close an ad-hoc route and release its ACTIVE lock. Pass `<plan.md> [verdict summary]` or `adhoc <

  • /apex-dispatch:report

    Print apex-dispatch's ledger report for a plan or a state directory — chain status, routes by class/tier/provider/mode, spawns, verdicts, escalations, tokens, estimated USD and the unverified-usage bucket; --compare for routed vs baseline, --decision for the decision layer vs routing. Pass a plan pa

  • /apex-dispatch:route

    Route one plan task or an ad-hoc ask through apex-dispatch's route.sh and print the ROUTE block (class, tier, model, effort, provider, roster, fan-out, review shape, budgets). Pass `plan <PLAN> --line N` or `adhoc --tags <csv> [--paths <globs>] [--acceptance <cmd>]` as $ARGUMENTS; add --dry-run to p

  • /apex-dispatch:run

    Drive one routed task end to end with apex-dispatch and apex-scope-loop — iterate → route → dispatch exactly the route → gate → risk tier → review shape → review → complete (or fail and escalate). Pass a plan path as $ARGUMENTS (or `adhoc <ROUTE_ID>` for an ad-hoc route made with /apex-dispatch:rout

Subagents (10)

  • adversarial-reviewer

    Read-only refutation pass for Tier C reviews. Tries to break the approving reviewers' conclusions with concrete inputs. Has no shell and no edit tools.

  • builder-high

    Builds one routed task inside the plan worktree and commits it, confined to the task's owned paths. Dispatched by apex-dispatch exactly as routed. Effort high variant, used by the escalation ladder.

  • builder-xhigh

    Builds one routed task inside the plan worktree and commits it, confined to the task's owned paths. Dispatched by apex-dispatch exactly as routed. Effort xhigh variant, used by the escalation ladder.

  • builder

    Builds one routed task inside the plan worktree and commits it, confined to the task's owned paths. Dispatched by apex-dispatch exactly as routed.

  • diagnoser

    Read-only diagnoser called on the second failure of a task. Explains why the previous attempts failed so the next builder starts from a cause, not a symptom.

  • docs

    Writes or updates documentation for one routed docs task, confined to the task's owned paths.

  • provider-runner

    Runs one external worker shim for a routed task and reports its result. Calls only the bin/worker scripts; never edits files.

  • researcher

    Read-only researcher that answers one routed question from the repository and the web, without changing anything.

  • reviewer

    Independent read-only reviewer for one routed task. Grades the exact head SHA from the brief against the Acceptance line and ends with one VERDICT line. Has no shell and no edit tools.

  • tester

    Writes or repairs tests for one routed task inside the plan worktree, confined to the task's owned paths.

Hooks (10)

  • PostToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/post-agent.sh"

  • PostToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/post-bash-prune.sh"

  • PostToolUseFailure

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/post-agent.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pre-agent.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pre-bash.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pre-edit.sh"

  • PreToolUse

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/pre-mcp.sh"

  • Stop

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/stop-gate.sh"

  • SubagentStart

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/subagent-start.sh"

  • SubagentStop

    bash "${CLAUDE_PLUGIN_ROOT}/hooks/subagent-stop.sh"

View source on GitHub

More from Skobyn/Apex-Dev-Skills

All 12 plugins →

Typed, probability-carrying answers for the questions apex-dispatch and apex-scope-loop leave to judgment (task class, risk tier): one CLI, rubric files, a fail-closed validator for every backend, tri-state uncertainty, and calibration that only a measurement job can grant. Off until a repo opts in; absent, slow or unconfigured, routing stays table-only.

1 skill · 5 commands · hooks

1

Routing and gate engine for apex-app: lanes + surface-ledger routing, cross-platform guardrails, and phases that close on computed obligations.

2 skills · 4 commands · hooks

1

Decide -> plan -> loop: SPARC-shaped ADR + phased plan authoring with scope sizing (bugfix, poc, feature) and an intent round, autonomous /loop + /schedule execution, per-phase swarm dispatch, The Gibson's harness gates (baseline green gate, independent review, Tier-C human approval), and an operate phase that watches landed plans against their declared checks.

2 skills · 3 commands · 2 agents

1

Trace, replay, and cost-attribute multi-agent runs: hook-driven JSONL AgentTrace, deterministic OrchestrationReplay, and per-subagent TokenLens — pure artifacts, with an optional read-only MCP dashboard server.

3 skills · 1 command · 1 MCP server · hooks

1

Interview-first greenfield project initializer: opens with 'what are you trying to build?', researches the context, then scaffolds a fully-governed repo (paved-road defaults), provisions the Apex-Dev-Skills suite (plus optional ruflo), and verifies it before the first commit. Generated projects are steered to build features via apex-plan/apex-loop. Supports TypeScript/JavaScript, Python, Go, and Rust.

3 skills · 1 command · 1 agent

1

Tool-contract validation + runtime reliability: a PostToolUse ledger that infers per-tool schemas and flags contract drift, plus FlakeGuard detection of non-deterministic agent failures (retries, partial failures, flaky tools).

2 skills · 1 command · hooks

1

Similar plugins

superpowers

claudecodex

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

15 skills · hooks

obra/superpowers 316,836

CMEM Pro

claude

CMEM Pro for the Claude app - hooks stream tool use to cmem.ai and inject observations into new sessions and agents

3 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

context7

claudecodex

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

upstash/context7 64,709

mempalace

claudecodex

AI memory system — mine projects and conversations into a searchable palace. 47 MCP tools, auto-save hooks, guided setup.

3 skills · 6 commands · 1 MCP server · hooks

MemPalace/mempalace 60,536

Build powerful, stateful CLI interfaces for any GUI application using the cli-anything harness methodology.

5 commands

HKUDS/CLI-Anything 58,665

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

Maintain apex-dispatch?

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.

claude & codex plugins12 plugins