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/AhmedHabiba/architorWrote 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/ahmedhabiba/architor/refine)<a href="https://agentmods.dev/commands/ahmedhabiba/architor/refine"><img src="https://agentmods.dev/badge/commands/ahmedhabiba/architor/refine/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/ahmedhabiba/architor/refine"><img src="https://agentmods.dev/badge/commands/ahmedhabiba/architor/refine.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 | $0.00000 | $0.00504 |
| Opus 5.5 | $0.00000 | $0.00202 |
| Sonnet 5.5 | $0.00000 | $0.00101 |
| Haiku 4.5 | $0.00000 | $0.00050 |
Grade A, and why
refine 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 2mo 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.
What it actually says
Read .arch/state.json first.
GATE CHECK: Something must be in awaiting_acceptance status. If nothing is pending, say: "Nothing is currently pending refinement. Use /status to check state."
The user wants to refine the current proposal. Their feedback is in $ARGUMENTS.
Determine what is being refined
If Phase 1 (Evaluation):
- Re-read
.arch/phase1-evaluation.md - Apply the user's feedback
- Re-analyze the specific areas mentioned
- Rewrite
.arch/phase1-evaluation.mdwith updates - Append refinement to
.arch/decisions.md - Present updated analysis, highlight what changed
If Phase 2 (Methodology):
- Re-read
.arch/phase2-methodology.mdand.arch/phase2-components-overview.md - Apply feedback to the OVERALL approach (not individual components)
- If user tries to refine a specific component's details: "Individual component refinement happens in Phase 3. In Phase 2, I can adjust the overall architecture pattern, component boundaries, or integration approach. What would you like to change at the system level?"
- Rewrite both files with updates
- Present updated methodology and component overview, highlight what changed
If Phase 3 (Component):
- Read the current component file from
.arch/components/[current_component].md - Apply feedback to this specific component only
- Check if changes affect previously accepted components — if yes, FLAG it: "⚠️ This change affects [Component X] which was already accepted. Specifically: [what breaks]. We should address this before proceeding."
- Rewrite the component file
- Present updated design, highlight what changed
After refinement:
- Status remains
awaiting_acceptance - Append to
.arch/decisions.md:### [DEC-NNN] Phase X | Refinement - **Decision:** Refined [target] — [brief description of what changed] - **Rationale:** [user's feedback that drove the refinement] - **Alternatives:** Previous version: [what was changed from] - **Trade-offs:** [any trade-offs introduced by the refinement] - **Risk:** [any new risk from the change] - **References:** [DEC-NNN of the original proposal being refined] - **Date:** [timestamp] - Ask for acceptance again
$ARGUMENTS
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.
- 2mo ago First seen · 47 lines · 0 tokens per session scan A 12131ac97652
refine is a command published in the GitHub repository AhmedHabiba/architor (6 stars, last pushed 7mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 504 tokens. 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.
refine compared with commands from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| refineAhmedHabiba/architor | Command | 6 | 7mo ago | MIT | A | 0 | Claude Code |
| learnReflexioAI/claude-smart | Command | 791 | 4d ago | Apache-2.0 | A | 14 | Claude Code |
| prototypegrainulation/grainulator | Command | 86 | yesterday | MIT | A | 0 | Claude Code |
| initazrtydxb/procoder | Command | 211 | 11d ago | Apache-2.0 | A | 16 | Claude Code |
| genshijin-compressInterfaceX-co-jp/genshijin | Command | 331 | 27d ago | MIT | A | 17 | Claude Code |
| responsive-design-specialistimsaif/design-with-claude | Command | 11 | 2d ago | MIT | A | 36 | Claude Code |
| brand-generateferdinandobons/brand-docs | Command | 291 | 4mo ago | MIT | A | 0 | Claude Code |