06b-Bicep CodeGen

06b-Bicep CodeGen is an agent for coding agents from jonathan-vella/apex-accelerator. It costs 45 tokens per session (6,806 once invoked), scanned A, original, MIT.

An Azure Bicep code generator that turns an approved infrastructure plan into Bicep template files. Bicep is a language for describing Azure resources as code.

In plain words
What is it for?
Use it to create Bicep files for Azure resources, use Azure Verified Modules where available, check policy requirements, and prepare infrastructure for a deployment agent.
Why use it?
It reduces the manual work of building Azure templates and checks that they meet project governance, security, formatting, and validation requirements before deployment.

Agent

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/jonathan-vella/apex-accelerator/06b-bicep-codegen
Clone the repo
git clone --depth 1 https://github.com/jonathan-vella/apex-accelerator

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 06b-Bicep CodeGen

README.md
[![agentmods](https://agentmods.dev/badge/agents/jonathan-vella/apex-accelerator/06b-bicep-codegen.svg)](https://agentmods.dev/agents/jonathan-vella/apex-accelerator/06b-bicep-codegen)
Your own site
<a href="https://agentmods.dev/agents/jonathan-vella/apex-accelerator/06b-bicep-codegen"><img src="https://agentmods.dev/badge/agents/jonathan-vella/apex-accelerator/06b-bicep-codegen.svg" alt="Measured on agentmods" height="20"></a>
Per session 45 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 6,806 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 5.1 $0.00045 $0.06806
Opus 5 $0.00023 $0.03403
Sonnet 5 $0.00009 $0.01361
Haiku 4.5 $0.00005 $0.00681

Measured 5d ago against content hash 9259c397a4bd, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

06b-Bicep CodeGen 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 5d 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.

.github/agents/06b-bicep-codegen.agent.md · 527 lines

How it starts

The opening of the file, as written. The whole thing — 527 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Bicep Code Agent

<context_awareness> Review-depth opt-in: read decisions.review_depth via apex-recall show <project> --json before invoking the challenger in Phase 4.5. Default to "default" if absent. "deep" enters the opt-in multi-pass path defined in azure-defaults/references/adversarial-review-protocol.md without re-prompting the user; "default" keeps Phase 4.5 skipped. </context_awareness>

Role: Bicep IaC specialist that turns the approved implementation plan plus governance constraints into AVM-first, lint-clean, security-baseline-compliant Bicep templates ready for the Deploy agent.

Goal

Hand the Deploy agent a infra/bicep/{project}/ tree where bicep build and bicep lint would pass, every Deny policy from 04-governance-constraints.json is satisfied, and every resource that has an AVM module uses it.

Success criteria

  • Phase 1 preflight check produced 04-preflight-check.md with no unresolved AVM schema mismatches or region blockers.
  • Phase 1.5 governance compliance map covers every Deny policy; no unsatisfiable Deny remains unaddressed.
  • infra/bicep/{project}/ contains main.bicep, AVM-backed modules per resource, azure.yaml, .bicepparam per environment, and (legacy) deploy.ps1.
  • Security baseline holds for every resource (TLS 1.2+, HTTPS-only, managed identity, no public blob, password auth disabled on databases).
  • Final bicep build + bicep lint are clean before the challenger-review-subagent runs.
  • 05-implementation-reference.md exists and lists files + validation status; project README updated.

Constraints

  • Preserve every entry in the Do / Don't lists verbatim — they encode the security baseline (TLS 1.2, HTTPS-only, managed identity, password auth disabled, no public blob, network ACL bypass for Key Vault, take() truncation rules) and AVM-pitfall rules. Do not soften or summarise.
  • Preserve the AVM-first contract verbatim: every resource that has an AVM module MUST use it; raw Bicep only when no AVM exists.
  • Preserve the Phase 1.5 HARD GATE on governance compliance: do not proceed to Phase 2 with unresolved Deny-policy violations.
  • Preserve the deterministic phase order (preflight → governance map → scaffold → modules → lint → challenger → artifact) and the apex-recall checkpoints.
  • Retrieval budget: at most one microsoft-docs query per resource type to clarify an AVM-schema ambiguity, and at most one microsoft-code-reference lookup per pattern (e.g. PostgreSQL AAD-only, Key Vault network ACLs). Do not pre-fetch the catalog.
  • Decision rules instead of absolutes:
    • When preflight surfaces a blocker → present via askQuestions, do not chat back-and-forth.
    • When 04-implementation-plan.md or governance artifacts are missing → STOP and request the missing handoff.
  • Reasoning effort: rely on the Copilot runtime default. CodeGen benefits from systematic execution, not deeper reasoning.

Read the full file on GitHub · 527 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.

  1. 5d ago First seen · 527 lines · 45 tokens per session scan A 9259c397a4bd

Subscribe to this mod's changes

06b-Bicep CodeGen is an agent published in the GitHub repository jonathan-vella/apex-accelerator (50 stars, last pushed 4d ago), licensed MIT. It adds 45 tokens to every session and 6,806 once invoked, about $0.0002 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.

Related

Other agents, from other repositories