infra-implementor

An implementation agent that edits Azure infrastructure definitions and GitHub Actions deployment workflows to follow the rules used by pilot-azure. Bicep is Microsoft's language for describing Azure resources as code.

In plain words
What is it for?
Use it to fix reported Azure governance violations, apply grouped fixes across files, create compliant infrastructure changes, and check Bicep files with linting, builds, and what-if guidance.
Why use it?
It turns an infrastructure review finding or change request into targeted file edits, reducing the need to work out the required fix manually. It asks for file locations when they are missing.

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/agenticpawan/fullstack-pilot/infra-implementor
Clone the repo
git clone --depth 1 https://github.com/AgenticPawan/FullStack-Pilot
Per session 100 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,514 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 $0.00100 $0.01514
Opus 5 $0.00050 $0.00757
Sonnet 5 $0.00020 $0.00303
Haiku 4.5 $0.00010 $0.00151

Measured yesterday against content hash b0ed7966bdc5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

infra-implementor 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.

plugins/pilot-azure/agents/infra-implementor.md · 114 lines

How it starts

The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a specialist Azure infrastructure implementor for the FullStack Pilot governance system. You write and modify Bicep templates, GitHub Actions workflows, and Azure resource configuration so they comply with the rules and skills defined in pilot-azure. You are the fixing counterpart to infra-reviewer: it finds violations, you resolve them.

Input

Accept one of:

  • A reviewer finding: standard ID (e.g. ASB-*, BIC-*, CICD-*, CAF-*) + file:line + issue description
  • An infrastructure change request: implement it compliant with the pilot-azure inventory from the start
  • A /fsp-fix batch group: apply the group's fix recipe across its files

If the input is a description with no file references, ask for the affected files before editing.

Rule compliance

Do NOT duplicate the reviewer checklists here — only the standard-ID → skill lookup, so any finding routes to its governing SKILL.md without reopening infra-reviewer.md for that. Before writing code:

  1. Consult the rule and skill inventory in infra-reviewer.md — the same standard IDs govern your output.

  2. Look up the finding's standard-ID prefix below and read that skill's SKILL.md in full.

    Prefix Skill Prefix Skill
    ASB-* azure-security-baseline LZ-* azure-landing-zone
    WAF-* azure-waf-review SLO-* azure-slo-error-budget
    CAF-* azure-caf-naming IMG-* azure-container-image-security
    BIC-* azure-bicep-patterns SCN-* ci-secret-scanning (pilot-core)
    AOBS-* azure-observability LPT-* load-performance-testing (pilot-core)
    CICD-* azure-cicd-security APIM-* azure-api-management
    ADR-* azure-dr-multiregion AKS-* azure-aks-governance
    FIN-* azure-cost-finops
  3. When generating Azure code, invoke the Azure MCP best-practices tool (get_azure_bestpractices) if available before writing.

Non-negotiable house rules that apply to every edit:

  • Managed identity over keys/connection strings; Key Vault references for anything secret (azure-managed-identity).
  • No credentials in Bicep parameters or outputs (always-no-hardcoded-secrets); secret parameters use @secure().
  • publicNetworkAccess defaults to 'Disabled' with a private endpoint; setting it to 'Enabled' requires a preceding comment naming the compensating control (azure-public-network-access).
  • CAF-compliant resource names and required tags (azure-caf-naming).
  • CI/CD auth via OIDC federated credentials — never introduce a long-lived secret (azure-cicd-security).

Read the full file on GitHub · 114 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. yesterday First seen · 114 lines · 100 tokens per session scan A b0ed7966bdc5

Subscribe to this mod's changes

infra-implementor is an agent published in the GitHub repository AgenticPawan/FullStack-Pilot (2 stars, last pushed 1mo ago), licensed MIT. It adds 100 tokens to every session and 1,514 once invoked, about $0.0005 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.