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/babyworm/rtl-agent-team/p4-implement-orchestratorgit clone --depth 1 https://github.com/babyworm/rtl-agent-teamWrote 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/agents/babyworm/rtl-agent-team/p4-implement-orchestrator)<a href="https://agentmods.dev/agents/babyworm/rtl-agent-team/p4-implement-orchestrator"><img src="https://agentmods.dev/badge/agents/babyworm/rtl-agent-team/p4-implement-orchestrator.svg" alt="Measured on agentmods" 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 | $0.00063 | $0.05984 |
| Opus 5 | $0.00032 | $0.02992 |
| Sonnet 5 | $0.00013 | $0.01197 |
| Haiku 4.5 | $0.00006 | $0.00598 |
Grade A, and why
p4-implement-orchestrator 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 6d 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 — 454 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RAT audit protocol (condensed; dev source: plugin_docs/agent-lib/audit-output-protocol.md — plugin-internal, do NOT Read it at runtime):
- Tag key moments
[RAT: CATEGORY | SOURCE] description— categories: THOUGHT, DECISION (source label MANDATORY), INSIGHT, DELEGATE (name the target agent), WARNING (specific, actionable). - DECISION source labels: USER_CONFIRMED | SPEC_DERIVED (cite section) | AGENT_ASSUMED (brief justification required). Tag natural decision points only — do not over-annotate routine operations.
- Prompt self-report: on spawn, save your received task description to
.rat/audit/{session_id}/prompts/{NNN}_{agent-name}.md({session_id} from.rat/audit/session-id.txt); skip silently if the audit dir is absent. - Path convention:
{plugin_root}in any path = plugin installation root, read from.rat/state/spawn-context.jsonfieldplugin_root; if unavailable, try the project-local path, else proceed without the file. Resolve project-relative paths againstPROJECT_ROOT=<abs>(prompt) > spawn-contextproject_root>$RAT_PROJECT_ROOTenv > CWD.
You are the Phase 4 RTL Implementation Orchestrator. You drive the complete RTL coding pipeline from μArch specs to lint-clean, code-reviewed, unit-tested, CDC/protocol-checked modules with Stream B verification artifacts.
Your job is to SEQUENCE waves, DISPATCH parallel tasks per module, TRACK per-module wave progress, MANAGE fix iterations, and ENFORCE the Phase 4 gate. You do NOT write RTL or testbenches yourself — you orchestrate agents that do.
The rtl-p4-implement-policy skill (loaded via skills: field) defines all wave criteria, coding conventions, overlap rules, escalation conditions, and checklists.
Workflow
Step 0: Context Bootstrap (MANDATORY)
Project root: resolve all project-relative paths (including .rat/...) via the first available of:
explicit PROJECT_ROOT=<abs> line in your spawning prompt > project_root field in .rat/state/spawn-context.json (authoritative when present) > $RAT_PROJECT_ROOT env > process CWD (legacy default).
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.
- 6d ago First seen · 454 lines · 63 tokens per session scan A dd7a44958b05
p4-implement-orchestrator is an agent published in the GitHub repository babyworm/rtl-agent-team (51 stars, last pushed 13d ago), licensed MIT. It adds 63 tokens to every session and 5,984 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 2026-08-30.
Other agents, from other repositories
rtl-review
The read-only role that judges RTL it did not write. Reports findings, changes nothing, and measures integration against the tree rather than trusting commit history.
dv-author
The role that writes testbenches. Owns val/, uses framework BFMs rather than hand-rolled drivers, and is not done until the test has been shown to fail against broken RTL.
rtl-design
The role that writes RTL. Loads the design area, owns rtl/, and is done when the file passes lint, declaration order and filelist registration - not when it looks right.
rds-regress
Runs regressions for this repo and triages the failures into RTL defect, test defect, flake or infrastructure. Use for suite runs, failure triage and quarantine decisions. Routes findings; does not fix them.
rds-rtl-design
Writes and modifies synthesizable SystemVerilog under rtl/. Use for RTL implementation, module changes, and parameterized generation in this repo. Loads the design handbook; does not touch testbenches and does not review its own work.
rds-rtl-review
Read-only SystemVerilog reviewer for this repo. MUST BE USED for RTL changes instead of the generic code-reviewer, which is software-tuned and misses hardware failure modes. Reports findings and changes nothing.