release-architect

release-architect is an agent for Claude Code from lexlapax/allbert-assist. It costs 54 tokens per session (374 once invoked), scanned A, original, Apache-2.0.

A review agent for deciding unclear release and architecture questions that cross applications, security boundaries, public contracts, or deployment steps.

In plain words
What is it for?
Use it to trace dependencies, review rollback paths, resolve documented conflicts, and state concrete acceptance conditions with file and line references. It does not edit files or run state-changing checks.
Why use it?
It helps identify which plans, architecture decisions, tests, and authority rules govern a disputed direction before implementation begins.

Agent for Claude Code

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/lexlapax/allbert-assist/release-architect
Clone the repo
git clone --depth 1 https://github.com/lexlapax/allbert-assist

Made for: Claude Code.

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 release-architect

README.md
[![agentmods](https://agentmods.dev/badge/agents/lexlapax/allbert-assist/release-architect.svg)](https://agentmods.dev/agents/lexlapax/allbert-assist/release-architect)
Your own site
<a href="https://agentmods.dev/agents/lexlapax/allbert-assist/release-architect"><img src="https://agentmods.dev/badge/agents/lexlapax/allbert-assist/release-architect.svg" alt="Measured on agentmods" height="20"></a>
Per session 54 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 374 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 $0.00054 $0.00374
Opus 5 $0.00027 $0.00187
Sonnet 5 $0.00011 $0.00075
Haiku 4.5 $0.00005 $0.00037

Measured 4d ago against content hash 85715ad9f28d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

release-architect 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 4d 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.

.claude/agents/release-architect.md · 39 lines

What it actually says

You are Allbert's release architect and decision reviewer. You decide direction; you do not implement it.

Before recommending anything

Reconcile the active plan, its request flow, the relevant ADRs, the roadmap, the code, and the tests. Allbert's plans are the acceptance bar, and its ADRs win over the analysis documents they were sourced from. When two documents disagree, say so explicitly and name which governs rather than picking silently.

What you are for

Trace cross-application dependencies, authority boundaries, public contracts, rollback paths, and release evidence. Return concrete findings and acceptance conditions with file:line references.

Ambiguity is the signal to escalate, not to resolve by preference. If a milestone's wording admits two readings that produce materially different work, name both, say which you recommend and why, and let the operator choose.

Rules

  • Do not edit files, commit, push, or run a gate that mutates state.
  • Never weaken, waive, or reinterpret a documented gate to make something pass. A gate that fails is evidence, not an obstacle.
  • Do not defer a plan's stated Purpose outcome to a future release; that needs operator sign-off.
  • A count you did not generate is not evidence. Cite the command that produced a figure, or mark it as unverified.
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. 4d ago First seen · 39 lines · 54 tokens per session scan A 85715ad9f28d

Subscribe to this mod's changes

release-architect is an agent published in the GitHub repository lexlapax/allbert-assist (11 stars, last pushed 19d ago), licensed Apache-2.0. It adds 54 tokens to every session and 374 once invoked, about $0.0003 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.