conduit

conduit is an agent for coding agents from 7xuanlu/boule. It costs 23 tokens per session (298 once invoked), scanned A, original, MIT.

A relay that forwards a prompt and its JSON response between the local agent and an external model command-line tool.

In plain words
What is it for?
It is for passing requests to Codex or agy, reading the returned JSON, and emitting it with only minimal syntax repair.
Why use it?
It keeps the external model’s view unchanged instead of having the relay reinterpret it.

Agent

Part of the boule plugin — 4 skills, 2 agents shipped together

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.

agentmods
npx agentmods add agents/7xuanlu/boule/conduit
Clone the repo
git clone --depth 1 https://github.com/7xuanlu/boule

Or install boule, the plugin that ships this one along with the rest of its 4 skills, 2 agents.

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.

agentmods badge for conduit

README.md
[![agentmods](https://agentmods.dev/badge/agents/7xuanlu/boule/conduit.svg)](https://agentmods.dev/agents/7xuanlu/boule/conduit)
Your own site
<a href="https://agentmods.dev/agents/7xuanlu/boule/conduit"><img src="https://agentmods.dev/badge/agents/7xuanlu/boule/conduit.svg" alt="Measured on agentmods" height="20"></a>
Per session 23 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 298 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00023 $0.00298
Opus 5 $0.00012 $0.00149
Sonnet 5 $0.00005 $0.00060
Haiku 4.5 $0.00002 $0.00030

Measured 3d ago against content hash ec3810b470ca, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

conduit 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 3d 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.

agents/conduit.md · 16 lines

What it actually says

You are a PASS-THROUGH CONDUIT to an external model's CLI, NOT a judge. Forward its view UNCHANGED. Do NOT re-evaluate, inject opinion, or editorialize.

Steps:

  1. Write the EXTERNAL PROMPT (everything below the marker, verbatim) to "$TMPDIR/council___in.txt".
  2. Run the provided command WITH THE BASH SANDBOX DISABLED (it needs network + IPC; a sandboxed attempt fails with "Operation not permitted"). codex runs in an isolated CODEX_HOME (auth-only) + neutral cwd and is hard read-only (-s read-only); agy runs from a neutral throwaway cwd in a terminal-restricted --sandbox — its print mode still auto-approves agy's own file-write tool, so this jails shell exec and relative writes but is weaker than codex's read-only.
  3. Read the model's final JSON (codex: from the -o out file; agy: from stdout).
  4. Emit that JSON VERBATIM as your structured output. Repair ONLY malformed syntax (unbalanced braces/quotes); NEVER change content. Ensure the "model" field is the requested id.

EXTERNAL PROMPT:

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.

  1. 3d ago First seen · 16 lines · 23 tokens per session scan A ec3810b470ca

Subscribe to this mod's changes

conduit is an agent published in the GitHub repository 7xuanlu/boule (1 stars, last pushed 2mo ago), licensed MIT. It adds 23 tokens to every session and 298 once invoked, about $0.0001 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 2026-08-31.