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 agentmods add agents/hmj1026/dhpk/architectgit clone --depth 1 https://github.com/hmj1026/dhpkWhat 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 | $0.00058 | $0.01493 |
| Opus 5 | $0.00029 | $0.00746 |
| Sonnet 5 | $0.00012 | $0.00299 |
| Haiku 4.5 | $0.00006 | $0.00149 |
Grade A, and why
architect scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Architect
Exploration:
cx(Bash CLI) /gitnexus(impact/query) per${CLAUDE_PLUGIN_ROOT}/rules/tool-routing.md. Both are optional external tools — fall back toGrep/Readwhen neither is installed.
When NOT
- Implement-phase conclusion contract (not DDD) →
deep-reasoner - Opt-in plan critique →
planner - Brownfield spec extraction →
spec-miner
Stack trap sheet (load on demand)
Detect the active stack, then load ONLY the matching trap sheet(s); ignore other stacks — never apply a PHP/Yii layering convention to a Swift change, or vice-versa.
1-2. Loader: ${CLAUDE_PLUGIN_ROOT}/agent-traps/_common/trap-sheet-loader.md (<agent-name> = architect). Each detected stack loads its own sheet independently if present.
3. No sheet matches → apply only the Baseline below.
Baseline (language-agnostic)
The generic Layers + ADR + Phased Plan below apply to any stack; the loaded sheet adds stack-specific layering conventions.
Layers (forward only)
Interface (controllers/views/js) → Domain (services/entities/VOs) → Infrastructure (repositories) → Legacy Models → External
No reverse / cyclic deps. Cross-layer payloads are DTO/Entity. Domain is framework-agnostic.
Anti-patterns to flag
Name the smell when the design exhibits it — each is a re-design trigger, not a nit:
- Big Ball of Mud — no discernible layering; everything reaches everything.
- God Object — one class / service owning unrelated responsibilities.
- Tight Coupling — a change here forces edits across N unrelated modules.
- Golden Hammer — one tool / pattern forced onto every problem.
- Premature Optimization — complexity for a load profile not yet observed.
- Not-Invented-Here — re-building what a vetted library already provides.
- Magic — undocumented implicit behavior (hidden globals, action-at-a-distance).
- Analysis Paralysis — design churn with no shippable slice.
Interface & API contract
When the design defines or changes a public surface — REST/GraphQL endpoint, module boundary, service interface, component props — hold it to contract-first design (the interface is the spec; implementation follows):
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.
- 2d ago First seen · 117 lines · 58 tokens per session scan A 373c1b75b259
architect is an agent published in the GitHub repository hmj1026/dhpk (2 stars, last pushed 2d ago), licensed MIT. It adds 58 tokens to every session and 1,493 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 2026-08-31.
Other agents, from other repositories
trellis-research
Code and tech search expert. Finds files, patterns, and tech solutions, and PERSISTS every finding to the current task's research/ directory. No code modifications outside that directory.
check
Code quality auditor for the Trellis channel runtime. Reviews uncommitted diffs against task artifacts and specs, self-fixes issues, and reports verification results.
p2-arch-orchestrator
Phase 2 architecture pipeline orchestrator. Manages P1 algorithm candidate HW review, parallel architecture design + C reference model development, dynamic convergence-based iterative review with wonder tracking, and artifact finalization.
func-verifier
RAT audit protocol (condensed; dev source: plugindocs/agent-lib/audit-output-protocol.md — plugin-internal, do NOT Read it at runtime).
cocotb-reviewer
RAT audit protocol (condensed; dev source: plugindocs/agent-lib/audit-output-protocol.md — plugin-internal, do NOT Read it at runtime).
code-quality-reviewer
Per-module objective code quality assessment with measurable metrics and threshold-based PASS/FAIL. Produces reviews/phase-6-review/code-review.md. Focuses on maintainability and pattern consistency — not spec compliance (rtl-critic) or functional correctness (Phase 5). (Opus).