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 skills/the01geek/prflow/create-issuenpx skills add The01Geek/prflow --skill create-issuegit clone --depth 1 https://github.com/The01Geek/prflowWhat 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 | $0.00121 | $0.06853 |
| Opus 5 | $0.00060 | $0.03426 |
| Sonnet 5 | $0.00024 | $0.01371 |
| Haiku 4.5 | $0.00012 | $0.00685 |
Grade A, and why
create-issue 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 — 350 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Completion checklist (do this first)
Before any of the pipeline's steps below, establish this seven-slot tracker.
Establish it by working down these rungs, moving on whenever one is unavailable: TodoWrite,
TaskCreate/TaskUpdate, update_plan, any candidate named in a runner-supplied listing of tools
it has not yet exposed, then a runner-advertised discovery mechanism — used to search for those
candidates, never as one itself. A candidate is unavailable when it is not exposed, when the
call returns a failure, or when it returns without a failure but confirms no tracker, through the
tool's own read-back where it exposes one or its rendered result otherwise. No rung ends the
run: hold a breadcrumb naming any failure and carry on to the next rung. Once every candidate is
exhausted, use the inline fallback in references/fallback-no-task-tool.md, loaded per the
Reference routing rules below — the route this skill takes when the runner exposes no task-tracking tool or the exposed one is disabled or unusable.
The seven slots:
- 1. Run Step 1's arm and write its evidence artifact
- 2. Clarify the user story until the Definition of Ready is met (Step 2)
- 3. Draft the issue and pass the no-options gate (Step 3)
- 4. Steelman the draft against the code, revise, re-pass the no-options gate, and append the steelman record to the derivation artifact (Step 3.5)
- 5. Bootstrap the audit state, offer the fresh-context audit, and run and act on any round the user elects (Step 3.6)
- 6. Present the rendered issue, get the user's explicit confirmation, then create it (Step 4, sub-steps 1–5)
- 7. After creation succeeds, run the gated implement-offer step — present the offer, or print the withheld-offer reason (Step 4, sub-step 6)
Mark each slot in_progress when you start it and completed only when done — the canonical status
vocabulary; a task tool whose status fields differ uses its nearest equivalents. The issue is
created only after the user explicitly confirms the rendered draft (slot 6) — never before. A
finished /prflow:docs-verify report is only slot 1.
What ships with it
27 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/audit-prompt-template.md 27 KB
- references/degradation-routing.md 11 KB
- references/fallback-audit-boundary-offer.md 5.9 KB
- references/fallback-audit-dispatch-arms.md 5.7 KB
- references/fallback-audit-evidence-degraded.md 3.0 KB
- references/fallback-audit-round-reconciliation.md 2.1 KB
- references/fallback-draft-write-recovery.md 2.4 KB
- references/fallback-implement-offer-tier-read.md 3.5 KB
- references/fallback-no-task-tool.md 3.7 KB
- references/fallback-read-only-sandbox.md 7.6 KB
- references/fallback-state-owner-unavailable.md 5.3 KB
- references/fallback-visual-specification.md 2.4 KB
- references/issue-template.md 37 KB
- references/quality-group-compatibility.md 3.5 KB
- references/quality-group-contracts.md 6.7 KB
- references/quality-group-premises.md 5.9 KB
- references/quality-group-regression.md 4.7 KB
- references/quality-group-semantic.md 2.2 KB
- references/quality-group-visual.md 1.6 KB
- references/revision-delta.md 6.2 KB
- references/step-2-clarify.md 29 KB
- references/step-3-5-steelman.md 34 KB
- references/step-3-6-audit-adjudication.md 27 KB
- references/step-3-6-audit-dispatch.md 19 KB
- references/step-3-6-audit-shared.md 19 KB
- references/step-3-6-audit.md 6.0 KB
- references/step-4-present-create.md 60 KB
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 · 350 lines · 121 tokens per session scan A 6e9c3cc69355
create-issue is a skill published in the GitHub repository The01Geek/prflow (115 stars, last pushed 3d ago), licensed MIT. It adds 121 tokens to every session and 6,853 once invoked, about $0.0006 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 skills, from other repositories
taiyi-plan
TaiyiForge /taiyi:plan — 项目规划入口:把 README/PRD 拆解为独立 change 清单,推荐 profile 和依赖关系。OpenCode / Claude / Codex / Cursor 通用。.
taiyi-task
TaiyiForge 第5阶段 — 任务拆解,TASK.md。四端通用。.
taiyi-ultrawork
TaiyiForge ultrawork — 并行切片 + Cursor Task 自动派发契约(对标 OMC ultrawork + spawnagent).
backlog
Read, work, and maintain a Git repo's deferred-work items in docs/backlog/, one file per item. Use when the user says "backlog", "check backlog", "what's on my backlog", "work the backlog", "address the backlog", "add to backlog", "clean up backlog", or when a review or task produced items that are real but not being…
triage
Work-unit triage for GitHub issues. Groups raw issues, fuses each group with the AGENTS.md northstar, and externalizes each routed unit to a substrate record a collaborator session is pointed at.
github-issue-manager
Classifies a user request as Bug, Compatibility, Feature, Question, or Task for the explyt/spring-plugin GitHub repo, lightly researches the relevant codebase, searches similar existing issues, and either reports an existing issue or creates a correctly titled and labeled one following the repository issue forms. Use…