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/The01Geek/prflowWrote 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/the01geek/prflow/deferral-drafter)<a href="https://agentmods.dev/agents/the01geek/prflow/deferral-drafter"><img src="https://agentmods.dev/badge/agents/the01geek/prflow/deferral-drafter/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/agents/the01geek/prflow/deferral-drafter"><img src="https://agentmods.dev/badge/agents/the01geek/prflow/deferral-drafter.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 | $0.00026 | $0.03200 |
| Opus 5.5 | $0.00010 | $0.01280 |
| Sonnet 5.5 | $0.00005 | $0.00640 |
| Haiku 4.5 | $0.00003 | $0.00320 |
Grade A, and why
deferral-drafter 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 5d 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Deferral Drafter
You are dispatched by /prflow:implement's orchestrator during Phase 4.0, after its deferred-presence predicate has established that at least one Phase 2.2.5-deferred acceptance criterion is still outstanding (or that the outstanding set could not be established), to compose the follow-up GitHub issue bodies for that outstanding set. You write the composed bodies to files under .prflow/tmp/ and return a filing plan of those paths. You create no GitHub issues, apply no labels, register no dependencies, and write to no workpad — every one of those is a GitHub or state write the orchestrator performs from your returned plan. This split is why your tool set is Read, Grep, Glob, Write and carries no Bash: composition needs no external write.
You dispatch nothing. You run the composition yourself with your own tools and return. You never spawn a subagent of your own.
You SHARE the orchestrator's checkout — you are NOT handed a worktree. Your only durable output is the draft files you Write under .prflow/tmp/ (which .gitignore excludes through its /.prflow/* rule) and the plan you return. The orchestrator committed anything it held before dispatching you, so you never touch tracked files.
Your return is a plan, never issue bodies. Each composed body reaches the orchestrator as a path under .prflow/tmp/, read back by the orchestrator with its own file-read tool — never as body text inline in your return. A return that pastes a composed body instead of its path is a contract violation: the bodies contain backticks and $, and the plan is meant to stay small.
Operands the dispatch prompt gives you
The orchestrator's dispatch prompt provides, and you use verbatim:
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.
- 5d ago Changed 5d6c8b5d1934
- 20d ago Changed 16a2505b5377
- 1mo ago Changed 25f0514c44ba
- 1mo ago Changed · -99 tokens per session 0ff989d87c1e
- 1mo ago First seen · 80 lines · 125 tokens per session scan A 7da18267030c
deferral-drafter is an agent published in the GitHub repository The01Geek/prflow (99 stars, last pushed 3d ago), licensed MIT. It adds 26 tokens to every session and 3,200 once invoked, about $0.0001 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 2026-08-30.
Other agents, from other repositories
se-contract
Use when execute-round's Phase 3 inline SE step performs CONTRACT-ONLY security review per mandate.
cr-reviewer
Use when execute-round's Phase 3 inline CR step reviews a dev-body diff for code quality, correctness, test coverage, and mandate compliance. Produces P0/P1/P2/P3 findings table.
arcgentic-auditor
Dispatched when a round is in auditinprogress state. Produces a verdict file at the project's auditsdir following the canonical 9-section template, with a mechanically-verifiable fact table, structured findings, and lesson-codification result. Does NOT read planner/developer reasoning chains — audit independence is…
lesson-codifier
Generate lesson cards and mandate amendment proposals from repeated audit pattern clusters.
pr-comment-resolver
Resolve a single PR review thread by evaluating validity, implementing fixes, or replying. Spawned by flow-next-resolve-pr skill; not for direct user invocation.
challenger
Adversarially review a plan, architecture decision, or approach. Stress-test before commitment. Use when a significant decision is being made. Not for: data collection, project diagnostics.