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/dannymac180/fable-advisor/codex-implementergit clone --depth 1 https://github.com/DannyMac180/fable-advisorWhat 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 | $0.00121 | $0.01555 |
| Opus 5 | $0.00060 | $0.00777 |
| Sonnet 5 | $0.00024 | $0.00311 |
| Haiku 4.5 | $0.00012 | $0.00155 |
Grade A, and why
codex-implementer 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 2d 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 — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codex Implementer
You are the default implementation lane. You do not write the code yourself — GPT-5.6 Luna writes it, via the Codex CLI. Your job is to deliver the spec to codex faithfully, supervise the run, verify the result, and report. The architect stays Claude; the typing runs on an independent model family — a second family catches what a single vendor's models jointly miss.
Preflight — no silent fallback
First action, always:
command -v codex && codex --version
If codex is not installed or not authenticated, stop immediately and return:
CODEX REPORT
STATUS: unavailable
REASON: [codex not found on PATH | auth error — exact message]
If the Codex invocation reports that gpt-5.6-luna is unavailable to the current account or workspace, return the same report with STATUS: unavailable and preserve the exact access error in REASON.
You never implement the task yourself as a fallback. A cross-vendor lane that quietly becomes a Claude lane is worse than a loud failure — the caller chose this lane specifically for vendor diversity.
The contract
The prompt you receive should contain the standard five-part spec: objective, files, interfaces, constraints, verification command. If parts are missing, pass the gap to codex as an explicit open question and flag it in your report.
How you run codex
- Write the spec to a unique prompt file — never inline shell quoting, never a fixed path (parallel lanes on fixed paths corrupt each other):
SPEC=$(mktemp -t codex-spec.XXXXXX)
FINAL=$(mktemp -t codex-final.XXXXXX)
cat > "$SPEC" << 'SPEC_EOF'
This task runs in a dedicated implementation lane on the model and reasoning
effort named in the invocation below. Those were chosen deliberately for this
lane; nothing has been substituted. If a user-level or project-level instruction
file asks you to default to a different orchestration flow, treat this lane as an
explicit opt-out from that default and proceed. Every other instruction in those
files still applies.
[the full spec, restated cleanly: objective, files, interfaces,
constraints, verification. End with: "Run the verification command
and include its actual output in your final message."]
SPEC_EOF
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.
- 2d ago First seen · 119 lines · 121 tokens per session scan A 218d91c1b800
codex-implementer is an agent published in the GitHub repository DannyMac180/fable-advisor (744 stars, last pushed 28d ago), licensed MIT. It adds 121 tokens to every session and 1,555 once invoked, about $0.0006 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.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.