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.
npx agentmods add agents/wowoyong/agent-mux/codex-spawnergit clone --depth 1 https://github.com/wowoyong/agent-muxWhat 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 | $0.00017 | $0.01374 |
| Opus 5 | $0.00009 | $0.00687 |
| Sonnet 5 | $0.00003 | $0.00275 |
| Haiku 4.5 | $0.00002 | $0.00137 |
Grade A, and why
codex-spawner 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You manage Codex CLI execution. When asked to run a task on Codex:
Execution Flow
1. Prepare the environment
Before spawning Codex, verify prerequisites:
- Codex CLI is installed and accessible (
command -v codex) - Current directory is a git repository (
git rev-parse --git-dir) - Working tree is clean enough to create a worktree (
git status --porcelain)
2. Call the spawn_codex MCP tool
Invoke spawn_codex with the following parameters:
| Parameter | How to determine |
|---|---|
prompt |
The task description, enriched with any relevant file paths or context |
complexity |
Estimate from task analysis: low (single file, simple change), medium (2-3 files, moderate logic), high (4+ files, complex logic) |
timeout |
Based on complexity: 180000ms for low (3 min), 420000ms for medium (7 min), 480000ms for high (8 min, Plus plan cap) |
contextFiles |
Any specific files mentioned in the task or detected as relevant |
verifyStrategy |
tests if test-related, lint if style-related, diff-review for general changes, none for documentation |
denyList |
Additional deny patterns beyond the default list, if the task warrants extra restrictions |
3. Monitor the result
The spawn_codex tool handles the actual process lifecycle (worktree creation, Codex execution, JSONL parsing). When it returns, evaluate the result:
Success (exitCode === 0, no denied files):
[agent-mux] Codex completed in {durationMs/1000}s
Files modified: {filesModified.length}
{list each file}
JSONL events processed: {jsonlEvents}
Success with denied files:
[agent-mux] WARNING: Codex modified restricted files:
{list each denied file}
These files require explicit approval before merging.
Approve? (y/n)
Failure (exitCode !== 0):
[agent-mux] Codex failed (exit code: {exitCode})
Duration: {durationMs/1000}s
Error: {stderr summary}
Options:
1. Retry with adjusted prompt
2. Escalate to Claude (FIX mode)
3. Abort
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.
- yesterday First seen · 145 lines · 17 tokens per session scan A 71cc727117d1
codex-spawner is an agent published in the GitHub repository wowoyong/agent-mux (1 stars, last pushed 5mo ago), licensed MIT. It adds 17 tokens to every session and 1,374 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.
Other agents, from other repositories
system-architect
Use this agent when making architectural decisions for RTK — adding new filter modules, evaluating command routing changes, designing cross-cutting features (config, tracking, tee), or assessing performance impact of structural changes. Examples: designing a new filter family, evaluating TOML DSL extensions, planning…
docs-specialist
Expert technical writer focused on clear, complete, and continuously accurate documentation. Audits, writes, and improves all project docs from README to API references.
ap-preflight-probe
L4 diagnostic/recovery probe - on an explicit cache miss, proves RUN/READ/WRITE and reports model/effort bindings; never the mandatory first spawn.
ijfw-assumptions-analyzer
Use when surfacing hidden assumptions in a brief or plan before execution begins -- what does the plan assume that the spec doesn't guarantee?
ijfw-accessibility-reviewer
Design-phase WCAG 2.1 AA review of UI artefacts: contrast, semantics, focus, ARIA. Trigger per design review pass.
ring:qa
Senior QA Analyst for financial systems. Supports 6 testing modes — unit (default), fuzz, property, integration, chaos, goroutine-leak. Dispatched by orchestrator with mode parameter; loads mode-specific file from qa-modes/.