A request router that sends plain-language tasks to different coding assistants, such as Claude Code, Cursor, Codex, Gemini, or Qwen. It can keep a session open for follow-up requests.

Use it to delegate coding or other agent tasks to a chosen assistant, select an assistant based on the request, and continue work in a session.

It removes the need to know which assistant command or connection method to use. It also follows an explicitly named assistant when the request includes one.

acp-router — Skill for Codex by XAIHT

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add XAIHT/Tlamatini --skill acp_router

Clone the repo

git clone --depth 1 https://github.com/XAIHT/Tlamatini

Made for: Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/xaiht/tlamatini/acp_router/github.svg)](https://agentmods.dev/skills/xaiht/tlamatini/acp_router)

Your own site

<a href="https://agentmods.dev/skills/xaiht/tlamatini/acp_router"><img src="https://agentmods.dev/badge/skills/xaiht/tlamatini/acp_router/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/xaiht/tlamatini/acp_router"><img src="https://agentmods.dev/badge/skills/xaiht/tlamatini/acp_router.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 64 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,257 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00064	$0.01257
Opus 5	$0.00032	$0.00629
Sonnet 5	$0.00013	$0.00251
Haiku 4.5	$0.00006	$0.00126

Measured 10d ago against content hash 77ab26c2ae79, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

acp-router scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Tlamatini/agent/skills_pkg/acp_router/SKILL.md · 99 lines

How it starts

The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ACP Router

Route a plain-language request to the right ACPX harness.

Decision rules

If the user's wording names a harness explicitly ("run this in Claude Code", "ask Cursor to..."), use that as harness directly.
If the user wants Codex chat conversation, prefer the native Codex binding (a separate skill); use ACPX Codex only when ACP//acp/acpx is named explicitly or when background spawn is needed.
Default mode to session so follow-up turns can be added with acp_send. Use one-shot only when the user asked for a single non-interactive task. (Note: mode is ignored by oneshot-prompt transport agents — claude / cursor / gemini / qwen / codex — because each turn is its own fresh process. Continuity inside the child is not preserved between acp_send calls on those harnesses.)

Transport awareness

The runtime now uses three transports. acp_doctor returns each agent's transport in details[].transport (when populated by the host) or you can infer from the agent_id:

oneshot-prompt (claude, cursor, gemini, qwen, codex) — re-spawns the CLI fresh per turn with the prompt as a CLI argument (claude -p "<task>", codex exec "<task>", ...) and captures stdout to EOF. acp_spawn runs the first turn synchronously and returns the answer in the events array; the assistant_message event holds the captured stdout.
json-acp (tlamatini self-host) — strict JSON envelope on stdin.
tui-repl (kiro, kimi, iflow, kilocode, opencode, pi, droid, copilot) — long-lived REPL over stdin. acp_spawn returns the session_id sub-second; harvest answers via acp_send_and_wait / acp_transcript.

Read the full file on GitHub · 99 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

10d ago First seen · 99 lines · 64 tokens per session scan A 77ab26c2ae79

Subscribe to this mod's changes

acp-router is a skill published in the GitHub repository XAIHT/Tlamatini (16 stars, last pushed yesterday), licensed MIT. It adds 64 tokens to every session and 1,257 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Building agents agent agentic-ai ai-assistant coding-assistant developer-tools django

copperhead

Change or verify a KiCad project through copperhead's gated pipeline. Use whenever a task touches .kicadsch or .kicadpcb files, a schematic, a PCB layout, a netlist, ERC/DRC, a BOM, or hardware design docs — instead of editing those files directly.

copperheadhq/copperhead · 69 tokens

commit

Inspect git status, draft a commit message that matches the repo's style, and commit staged changes.

mochow13/keen-code · 22 tokens

review

Review current code changes for bugs, regressions, test gaps, and project-guideline violations.

mochow13/keen-code · 21 tokens

tura

Work in the Tura agent-runtime repository. Use for Tura architecture, Rust backend, GUI/TUI, prompts, commands, providers, sessions, documentation, tests, packaging, and release work in this directory.

Tura-AI/tura · 46 tokens

grade-and-fix-daedalus

How to grade a Daedalus run (or pasted agent transcript), root-cause the failure, and ship the fix to Daedalus CORE as a stacked PR — without modifying the prompt-vault sandbox (read-only grading). Covers recurring bug archetypes (Unicode punctuation patch mismatch, syntax-vs-type mislabel, emoji box misalignment…

bgill55/daedalus · 130 tokens

add-slash-command

How to add a new slash command to Daedalus (src/commands), including the docs-sync step that breaks CI if skipped.

bgill55/daedalus · 32 tokens