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 skills/rijuvashisht/magneto/codexnpx skills add rijuvashisht/Magneto --skill codexgit clone --depth 1 https://github.com/rijuvashisht/MagnetoWrote 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.
[](https://agentmods.dev/skills/rijuvashisht/magneto/codex)<a href="https://agentmods.dev/skills/rijuvashisht/magneto/codex"><img src="https://agentmods.dev/badge/skills/rijuvashisht/magneto/codex.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00000 | $0.00858 |
| Opus 5 | $0.00000 | $0.00429 |
| Sonnet 5 | $0.00000 | $0.00172 |
| Haiku 4.5 | $0.00000 | $0.00086 |
Grade A, and why
codex 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/codex — Cross-Model Second Opinion with Convergence Scoring
Role: Cross-Model Reviewer
Inputs: current diff or target code
Outputs: review-report.md with multi-model analysis
Audit actions: codex.started, codex.convergence_scored, codex.completed
Why This Skill Exists — The Moat
Single-model review has blind spots. /codex runs the SAME diff through 2-3 different LLMs simultaneously and scores their convergence:
- 3/3 agree on a finding → high confidence, probably a real bug
- 2/3 agree → medium confidence, worth investigating
- 1/3 flagged uniquely → could be insight the others missed, or model-specific noise
gstack's /codex uses one OpenAI model. Magneto's /codex uses all configured runners.
Configuration
Runners are pulled from Magneto's existing runner registry. Default for /codex:
- OpenAI
gpt-4o(if OPENAI_API_KEY set) - Anthropic Claude (if ANTHROPIC_API_KEY set)
- Ollama local (if Ollama reachable)
Configure in .magneto/codex.config.json:
{
"runners": ["openai", "anthropic", "ollama"],
"timeout": 60000,
"costBudget": 0.50
}
Modes
magneto codex review # pass/fail gate
magneto codex challenge # adversarial: actively try to break the code
magneto codex consult # open-ended Q&A with cross-model input
The Protocol
Step 1: Build the review prompt
Load:
- The diff (or target files)
- Knowledge graph context (affected communities + god nodes)
- Memory patterns relevant to changed files
Single prompt, sent to all runners in parallel.
Step 2: Run in parallel
magneto run --runner openai --mode review < prompt.md > openai-review.json &
magneto run --runner anthropic --mode review < prompt.md > anthropic-review.json &
magneto run --runner ollama --mode review < prompt.md > ollama-review.json &
wait
Step 3: Score convergence
For each finding:
- Normalize across models (same file + line + category → same finding)
- Count how many models flagged it
- Tag as HIGH (3/3), MEDIUM (2/3), UNIQUE (1/3)
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.
- 6d ago First seen · 107 lines · 0 tokens per session scan A 250584c59b2d
codex is a skill published in the GitHub repository rijuvashisht/Magneto (6 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 858 tokens. 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 skills, from other repositories
bootstrap-repo-analysis
First-time analysis of a repository with no prior reviewer outcomes. Crawl historical merged-PR review feedback with the gh CLI (plus any preloaded samples), extract the team's review norms, and synthesize the initial per-repo review-style prompt. Use this for a cold-start repo; use continual-learning instead once the…
continual-learning
Nightly refinement of an existing per-repo review-style prompt using this reviewer's own finding outcomes. Read confirmed (resolved-by-commit / thumbs-up) and dismissed (thumbs-down) findings, promote the bug patterns the team actually fixes, demote the false-positive patterns, reconcile against the current prompt…
oracle
Oracle second-model review: bundle prompts/files, debug, refactor, design.
chrome-cdp
Drive a headless Chrome over the Chrome DevTools Protocol (CDP) for browser QA — navigate, click, fill forms, read the DOM/accessibility tree, screenshot, and assert. Use whenever a task requires loading a web page and interacting with it like a user. Chrome is launched by a bash step (recipe below); this skill…
coding-standards
Universal coding standards, best practices, and patterns for TypeScript, JavaScript, React, and Node.js development.
review-work
Quality gate: verify each acceptance criterion of a completed task/work unit, run quality checks, and create follow-up tasks for gaps. Use before merging or to audit delivered work. Invoked as /agiflow:review-work . Uses getworkunit, gettask, updatetask, createtask, createtaskcomment.