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.
npx skills add a1f/agent-templates --skill improve-architecturegit clone --depth 1 https://github.com/a1f/agent-templatesWrote 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/skills/a1f/agent-templates/improve-architecture)<a href="https://agentmods.dev/skills/a1f/agent-templates/improve-architecture"><img src="https://agentmods.dev/badge/skills/a1f/agent-templates/improve-architecture/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/skills/a1f/agent-templates/improve-architecture"><img src="https://agentmods.dev/badge/skills/a1f/agent-templates/improve-architecture.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 | $0.00051 | $0.02539 |
| Opus 5.5 | $0.00020 | $0.01016 |
| Sonnet 5.5 | $0.00010 | $0.00508 |
| Haiku 4.5 | $0.00005 | $0.00254 |
Grade A, and why
improve-architecture 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.
How it starts
The opening of the file, as written. The whole thing — 179 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Improve Architecture
Audits a subsystem and proposes one named operation per module, each earned — or killed — by a caller list re-traced blind.
Proposes, never edits — reviewer and critic judge a diff; this audits existing
code, seeding /breakdown or /make-pr on approval.
Rubric: ~/.claude/at/rules/design-principles.md (the canonical rubric the installer
composes into the state root, not the audited repo's own files) — defines deep vs shallow
and every red flag or principle below.
/improve-architecture [path/subsystem] [--write-issue]
1. Scope + inventory + load the rubric (ask if no path)
2. Sweep clusters, then a cross-cluster pass (read-only subagents)
3. Blind re-trace, verify, kill the weak (fresh skeptic per survivor)
4. Rank → findings.md, or report clean (gate: you pick)
5. Grill the winner, then hand it off (/breakdown or /make-pr)
The operations
Every finding is one operation anchored on a primary module (naming its counterpart for MERGE/MOVE/EXTRACT) — never "improve this"; the red flag is its evidence.
| Op | Propose when… | Red flag / principle |
|---|---|---|
| SPLIT | one module owns two jobs that don't share information — its name needs an "and", a flag parameter forks its behavior, or pure logic is tangled with I/O | Multi-concern · Flag parameter · Scattered side effects |
| MERGE | two modules are always used together, a thin layer only forwards, or several modules mirror execution order (read→modify→write) and each must know the others' format | Pass-through · Conjoined methods · Temporal decomposition |
| EXTRACT (add) | the same decision/format is coded in 3+ places, or behavior can't be swapped or tested without editing it in place (no seam) | Information leakage · Repetition · Special-general mixture |
| MOVE (co-locate) | code that changes together is scattered; gather it without merging interfaces | Change amplification |
| DELETE (remove) | a module's interface rivals its body and the code is trivial to inline at each caller | Shallow module · Pass-through |
| DEEPEN (reshape in place) | the job is right but the interface leaks complexity — overexposed options, an error forced onto every caller, or a name that implies behavior it lacks | Overexposure · Define errors out of existence · Non-obvious code |
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 · 179 lines · 51 tokens per session scan A 57ac27a8fe7c
improve-architecture is a skill published in the GitHub repository a1f/agent-templates (2 stars, last pushed 2mo ago), licensed MIT. It adds 51 tokens to every session and 2,539 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…