메뉴
Optimize multi-agent coding CLI stacks for lower token burn across Claude Code, Codex, Kimi Code CLI, and Hermes; includes workflows for unknown agents, cross-agent updates, MCP/profile propagation, and usage/provider rotation analysis.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Create, wire, and maintain 33god/DeLoNET projects. Covers pjangler/CommonProject bootstrap, Hermes PM and scrum-master/Ticket Sentinel provisioning with inherited profile config, mandatory mise/.env.op wiring, BMAD, Hindsight/Bloodbank hooks, and shared Hermes fleet updates. Use when running `pjangler init`, `pjangler hermes-agent`, or `mise run init-project`; adding a PM or ticket sentinel; updating Hermes for the fleet; changing the fleet default model/config; touching hermes-agent-template or pjangler provisioning; wiring mise/op inject; installing BMAD; or configuring agent hooks. Keywords: pjangler, CommonProject, hermes-agent-template, .project.json, Ticket Sentinel, inherited profile config, Hermes update, BMAD, Hindsight, Bloodbank.
Keep one hand-edited master config and generate per-agent CLI configs from it. Covers SSOT fan-out for agent hooks and skills: master-to-dialect propagation, ambiguity lock files, drift checks, and the Bloodbank services/agent-hooks reference implementation. Use for hooks.master.json, hooks.mappings.lock.json, generated-config drift, fan-out, SSOT, agent hooks, defer_to_global, .agents/local.json, sync.py, project-scoped hooks, skill fan-out, and new agent CLI dialects. Do NOT use for using pjangler to create projects or adoption checklists (33god-projects), event schemas or Bloodbank topology (bloodbank-integration), versioning (mise-versioning), or single-target config.
Use when a task needs an isolated hidden Linux desktop or workspace-owned browser: GUI app QA, web/browser/shopping automation, sandboxed app observation, or stale workspace cleanup. Routes agent-workspace-linux MCP tools on demand. Does NOT apply to host desktop/Chrome control, generic MCP setup, or pure code/file edits.
Create a document using the Client Profile Sheet template and its retained reference file. Use when the user selects this template, names Client Profile Sheet, or explicitly invokes $artifact-template-client-profile-sheet. Create an evidence-first prospect or client profile with qualification, source confidence, systems, stakeholders, opportunity boundaries, risks, next actions, and engagement history.
Codifies Jarad DeLorenzo's voice, snark, humor, moral compass, ideals, preferred structure, and human idiosyncracies into a set of workflows and tools. Use when asked to write a blog post, or when you're just looking for a way to express yourself.
Excavates lost feature context from git history, docs, tickets, PRs, and session breadcrumbs to restore missing project memory to Hindsight. Use when the user says a feature has regressed/vanished/been forgotten, summons GitArcheologist, or asks to dig up the remains of past work.
| name | agent-token-efficiency |
| version | 0.1.0 |
| description | Optimize multi-agent coding CLI stacks for lower token burn across Claude Code, Codex, Kimi Code CLI, and Hermes; includes workflows for unknown agents, cross-agent updates, MCP/profile propagation, and usage/provider rotation analysis. |
| pipeline-status | new |
Use this skill when the user wants to save tokens, reduce context bloat, configure coding agents, propagate skills/tools/profiles/MCP servers, analyze usage, or design provider/model rotation for continuous agentic coding.
This skill codifies a layered strategy:
Run the bundled Typer CLI:
python scripts/agent_efficiency.py --help
Recommended install from this skill directory:
python -m pip install -e .
aeff --help
Given the path to an unknown agent CLI or repository, inspect its capabilities, discover docs hints, map supported strategies from Claude/Codex/Kimi/Hermes, and produce/apply a token-efficiency optimization plan.
aeff optimize-unknown /path/to/agent-or-executable --docs-url https://example.com/docs --out ./reports
Use --apply only after reviewing the report:
aeff optimize-unknown /path/to/agent-or-executable --apply
The workflow looks for:
It then applies equivalent strategies where safe:
Apply one update across the supported CLI fleet.
aeff propagate-update --kind mcp --name context7 --source assets/mcp/context7.json --target all
Supported --kind values:
mcpskillprofileagentinstructionconfigDry-run is default. Add --apply to mutate files or run agent CLI commands.
Read local usage metrics, inspect configured agents/providers, compare provider free/throttled capacity, and calculate a rotation plan.
aeff analyze-usage --providers assets/configs/providers.yml --desired-rpm 4 --desired-daily-requests 500
With web refresh of provider notes:
aeff analyze-usage --allow-web --providers assets/configs/providers.yml
This workflow should produce:
scripts/agent_efficiency.py — Typer CLI implementation.assets/configs/providers.yml — provider quota and rotation assumptions.assets/mcp/*.json — bundled MCP server definitions.assets/agents/* — reusable scout/reviewer/research agents.assets/templates/* — AGENTS.md, CLAUDE.md, and reports.references/*.md — implementation playbooks.When this skill is active:
aef doctor or python scripts/agent_efficiency.py doctor before modifying a machine.~/.agent-token-efficiency/reports or the requested output directory.