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.
git clone --depth 1 https://github.com/navraj007in/architecture-cowork-pluginWrote 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/navraj007in/architecture-cowork-plugin/check-state)<a href="https://agentmods.dev/commands/navraj007in/architecture-cowork-plugin/check-state"><img src="https://agentmods.dev/badge/commands/navraj007in/architecture-cowork-plugin/check-state/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/navraj007in/architecture-cowork-plugin/check-state"><img src="https://agentmods.dev/badge/commands/navraj007in/architecture-cowork-plugin/check-state.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 | $0.00016 | $0.02327 |
| Opus 5 | $0.00008 | $0.01163 |
| Sonnet 5 | $0.00003 | $0.00465 |
| Haiku 4.5 | $0.00002 | $0.00233 |
Grade A, and why
check-state 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 12d 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 — 258 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/architect:check-state
Validates _state.json for schema compliance, type errors, and per-command write rule violations. Identifies stale fields, missing required sections, and inconsistencies across state.
Trigger
/architect:check-state
/architect:check-state [--fix] # auto-fix simple issues
/architect:check-state [--strict] # fail on warnings (not just errors)
Purpose
The _state.json file accumulates output from 50+ commands, each owning specific fields. Over time, inconsistencies emerge:
- Field values that don't match their declared type (e.g.,
project.stageis array instead of enum) - Stale fields (set by command A, never updated, contradicts current command B output)
- Missing required fields for the project's stage
- Violations of per-command write rules (command X overwrote field owned by command Y)
This command scans the state file and reports all issues with severity levels.
Input
No required input. Reads architecture-output/_state.json in the current project.
Output Files
1. architecture-output/state-check-report.md (always generated)
# State Validation Report
**Date:** 2026-04-24 14:32:10
**Project:** my-startup
**Stage:** growth
## Summary
✅ **Pass Rate:** 87/92 checks passed (94%)
❌ **Critical Errors:** 2
⚠️ **Warnings:** 3
ℹ️ **Info:** 0
---
## Critical Issues
### 1. Invalid type: `project.stage` (CRITICAL)
**Field:** `project.stage`
**Expected:** `"concept" | "mvp" | "growth" | "enterprise"`
**Actual:** `["mvp", "growth"]` (array, not string!)
**Consequence:** Downstream commands interpret stage as growth only; this breaks cost estimation and roadmap generation.
**Fix:** `jq '.project.stage = "growth"' _state.json > _state.json.fixed && mv _state.json.fixed _state.json`
### 2. Stale field: `blueprint.deepen_passes` (CRITICAL)
**Field:** `blueprint.deepen_passes`
**Value:** `0` (set 30 days ago)
**Problem:** Current blueprint shows `deepen_passes: 3`, but state shows `0`. This contradicts the actual output.
**Consequence:** Next blueprint run may not deepen correctly.
**Fix:** Update to `3` or delete (will be regenerated by next blueprint run).
**Command:** `jq '.blueprint.deepen_passes = 3' _state.json > _state.json.fixed && mv _state.json.fixed _state.json`
---
## Warnings
### 1. Write-rule violation: `design` field (WARNING)
**Field:** `design`
**Owned by:** `design-system` command
**Last modified:** 2026-04-20 (blueprint ran on 2026-04-20, may have overwritten design tokens)
**Consequence:** Design is out of sync with actual design-system output.
**Suggested fix:** Run `/architect:design-system` to refresh design tokens.
### 2. Missing recommended field: `entities` (WARNING)
**Stage:** growth
**Recommendation:** growth-stage projects should have `entities` defined (currently absent).
**Consequence:** Data-model generation will fall back to SDL entities only (may miss business logic).
**Suggested fix:** Run `/architect:generate-data-model` to populate entities.
### 3. Unused field: `cost_estimate.old_currency_format` (WARNING)
**Field:** `cost_estimate.old_currency_format`
**Found:** This field is not in the canonical schema and may be stale.
**Consequence:** Ignored by all commands.
**Suggested fix:** Delete via `jq 'del(.cost_estimate.old_currency_format)'`
---
## Schema Validation Passed
✅ All top-level fields are correct types
✅ All project stage requirements met (growth stage)
✅ All command-owned field write rules respected (no overlaps)
✅ Design colors are valid hex (6 verified)
✅ Component types are recognized (8 components)
✅ Entity schema looks good (12 entities, all with types)
---
## Recommendations
**Priority 1 (run immediately):**
1. Fix `project.stage` type (array → string)
2. Reconcile `blueprint.deepen_passes` (0 → 3)
**Priority 2 (run before next major command):**
3. Run `/architect:design-system` to refresh design tokens
4. Run `/architect:generate-data-model` to populate entities
**Priority 3 (clean-up, can wait):**
5. Remove unused `cost_estimate.old_currency_format` field
---
## Fix Instructions
**Auto-fix (safe issues only):**
```bash
/architect:check-state --fix
This will:
- Correct type mismatches (array → string, etc.)
- Remove unknown fields
- Preserve all data
- Create backup before modifying
Manual fixes:
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.
- 12d ago First seen · 258 lines · 16 tokens per session scan A c1f8e7dfb07b
check-state is a command published in the GitHub repository navraj007in/architecture-cowork-plugin (2 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 16 tokens to every session and 2,327 once invoked, about $0.0001 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.
Other commands, from other repositories
nyann:retrofit
Audit an existing repo against a profile and fix what's drifted. Unlike doctor (read-only), retrofit detects missing hooks, misconfigured gitignore, documentation gaps, and non-compliant history, then offers to remediate via bootstrap. Idempotent — safe to re-run.
nyann:diagnose
Bundle a redacted support-grade snapshot of the current nyann state (state + drift + hooks + config + prereqs) for inclusion in a bug report.
spec-bugfix
Start a new bugfix spec with structured bug analysis, design, and tasks.
zoom-out
Zoom out and provide a high-level map of modules and callers for unfamiliar code.
analyze
Analyze a .NET project or solution — list projects, check diagnostics, and present a structured summary.
duck-off
Turn off Rubber Duck mode and answer normally.