Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add Hainrixz/the-architect/plugin install the-architectWrote 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/commands/hainrixz/the-architect/architect-next)<a href="https://agentmods.dev/commands/hainrixz/the-architect/architect-next"><img src="https://agentmods.dev/badge/commands/hainrixz/the-architect/architect-next/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/hainrixz/the-architect/architect-next"><img src="https://agentmods.dev/badge/commands/hainrixz/the-architect/architect-next.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00066 | $0.01713 |
| Opus 5.5 | $0.00026 | $0.00685 |
| Sonnet 5.5 | $0.00013 | $0.00343 |
| Haiku 4.5 | $0.00007 | $0.00171 |
Grade A, and why
architect-next 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 1mo 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/architect-next — the resume protocol
You are being invoked as /architect-next. A build that spans sessions needs one question answered
in a fresh context with no memory: what do I do next, and how do I know when it is done?
This command answers it from tasks.json. Nothing else. No interview, no redesign.
Field names are defined by ${CLAUDE_PLUGIN_ROOT}/templates/tasks-schema.md — read it first and
use its names verbatim. If a bundle's tasks.json disagrees with the schema, report the mismatch
instead of guessing what a field meant.
Locate the bundle
| Argument | Do |
|---|---|
| A directory | Read <dir>/tasks.json. |
A tasks.json path |
Read it. |
| Empty | Glob ./blueprints/*/tasks.json, then ./tasks.json. One match → use it. Several → list them and ask. |
| Nothing found | Say so and point at /architect (new project) or /architect-brownfield (existing repo). Do not invent a task list. |
Verify commands run from the target project root — the repo being built — not from the bundle
directory. Every command in a task's verify array is written to be runnable from that root. If
the bundle is not inside the project it builds (a design session that produced
./blueprints/<slug>/ in some other directory), say so and stop: running verify from the wrong
working directory produces failures that have nothing to do with the code, and a red result nobody
can trust is worse than no result. Tell the user to move the bundle into the project, or to re-run
/architect-next <path-to-bundle> from the project root.
Handle an interrupted task first
Before selecting anything, look for a task whose status is in_progress. That is not a ready task —
it is crash recovery. A previous session wrote in_progress before touching code and never got
to finish, so the working tree may already be half-changed.
Exactly one such task can exist. When you find it:
- Print its card, labelled
RESUMING. - Run every command in its
verifyarray. All pass → set its status todoneand continue to selection. Any fail → this is the task to work on; report which commands failed and stop. - Never hand out a different task while one is
in_progress. Re-handing an interrupted task out as fresh work is how a build ends up applying the same migration twice.
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.
- 1mo ago First seen · 130 lines · 66 tokens per session scan A 501729d1bf8a
architect-next is a command published in the GitHub repository Hainrixz/the-architect (561 stars, last pushed 5d ago), licensed MIT. It adds 66 tokens to every session and 1,713 once invoked, about $0.0003 per session on Opus 5.5. 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 commands, from other repositories
status
Display project status, active tracks, and next actions.
amend-plan
Plan Amend - Discuss and apply amendments to a plan.json implementation plan.
update-counts
Explicitly refresh public toolkit and committed marketplace source snapshots and numeric documentation.
sdd
A command that creates three connected planning files for a software feature: requirements, design, and tasks.
scout-meta-review
Run an interactive Scout meta-review — a system-level audit that sits above the individual session types. Checks whether sessions are running, mistake audit is trending well, proposals are flowing, KB files are healthy, and data-source coverage is consistent across session types. Runs in the current conversation.
triage
Process the user-feedback queue — triage new reports, keep the status mirror honest, and (at release time only) send the shipped emails.