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 commands/heznpc/ploidy/architecturegit clone --depth 1 https://github.com/heznpc/PLOIDYWhat 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 | $0.00033 | $0.00874 |
| Opus 5 | $0.00016 | $0.00437 |
| Sonnet 5 | $0.00007 | $0.00175 |
| Haiku 4.5 | $0.00003 | $0.00087 |
Grade A, and why
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 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 — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The user is weighing an architecture-level change:
$ARGUMENTS
You hold the existing system's context (modules, prior decisions, fail modes, operational history). That is precisely the bias the proposal is competing against — Ploidy's paper shows deep-context reviewers anchor on "fits the current shape" over "is actually correct". Run the four-step eval without asking to confirm.
1 · Write the Deep-context take
If $ARGUMENTS is a filepath (ends in .md, .txt, contains /),
read that file first via the Read tool. Otherwise treat it as the
proposal text.
In ~300 words:
- Fit — how does the proposal interact with the current system? Name the modules it touches and the abstractions it respects or violates.
- Concrete risks — what breaks, what gets slower, what becomes harder to test / deploy / observe?
- Sunk-cost check — list what in your current system you find yourself defending here. For each, state whether the defense is "this abstraction is load-bearing" or "we just spent a lot building it". If nothing, say so explicitly.
- Non-obvious alternatives — is there a version of the proposal that solves the same problem with less change? One sentence each.
Tag every concern HIGH / MEDIUM / LOW confidence.
2 · Spawn a Fresh sub-agent
Use the Agent tool (subagent_type="general-purpose"). It must see
only the proposal text, no repo knowledge.
Prompt the subagent with exactly:
You are evaluating an architecture proposal in a codebase you have never seen. Only the proposal text is available:
<PROPOSAL TEXT — paste the expanded $ARGUMENTS here>
Answer in under 250 words:
- Stated in your own words, what problem is this trying to solve? If the problem is unclear, say so — do not guess.
- What assumptions is the proposal making about the existing system? List each; mark ones that sound uncommon.
- What are generic architectural alternatives (event-driven, batch, read-model, CQRS, etc.) that the proposal does not mention? One sentence each.
- Where is the proposal most likely to be overcomplicating? Be specific about which piece.
Do not ask for repo links. Do not request more context.
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 · 103 lines · 33 tokens per session scan A e19bb87a0ee2
architecture is a command published in the GitHub repository heznpc/PLOIDY (0 stars, last pushed 6d ago), licensed MIT. It adds 33 tokens to every session and 874 once invoked, about $0.0002 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
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.