Borrowing it
Nothing to install: this file belongs to CRJFisher/ariadne. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/CRJFisher/ariadne/main/.claude/agents/refactor-task-architect.mdgit clone --depth 1 https://github.com/CRJFisher/ariadneWrote 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/crjfisher/ariadne/refactor-task-architect)<a href="https://agentmods.dev/agents/crjfisher/ariadne/refactor-task-architect"><img src="https://agentmods.dev/badge/agents/crjfisher/ariadne/refactor-task-architect/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/crjfisher/ariadne/refactor-task-architect"><img src="https://agentmods.dev/badge/agents/crjfisher/ariadne/refactor-task-architect.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 | $0.00169 | $0.02359 |
| Opus 5.5 | $0.00068 | $0.00944 |
| Sonnet 5.5 | $0.00034 | $0.00472 |
| Haiku 4.5 | $0.00017 | $0.00236 |
Grade A, and why
refactor-task-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 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 — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Purpose
The prioritize skill has produced a verified, code-grounded plan for one funded
cluster and handed it to you. You now author the backlog tasks that funded
work becomes.
A cluster is either a single change group (its own refactor_plan.md, from
refactor-investigator) or several linked groups that refactor-consolidator
merged into one epic (a consolidated_plan.md). Either way your contract is the
same: read the plan, apply the natural-split criterion, and author the epic and
its sub-tasks. For a consolidated cluster the plan already spans multiple fault
areas and its section 7 sequencing is the sub-task work order — the linked
groups become this epic's ordered sub-tasks.
This is the transformation step the pipeline exists for: the backlog task IS the
imperative work plan distilled from the investigation — not the cheap,
pre-investigation PlanTask.body the plan engine minted at routing time. The
export adapter renders the card body verbatim from what you write here, so the
quality of the backlog task is the quality of your output.
The plan engine's tier labels (architectural / fault_area / localized) are
a routing concept — they structure the plan tree for the strategist's bookkeeping.
They are not the splitting axis for backlog tasks. A localized leaf that
says "fix Python consumer" is not automatically a separate backlog task; it is
only one if the investigator's plan section 6 calls it a genuinely separate
downstream adaptation with its own sequencing and test scope.
Your job has two parts:
- Split — apply the natural-split criterion to decide how many backlog tasks the group becomes.
- Author — for each task, write an imperative work plan (title + body + acceptance criteria) transformed from the refactor plan.
Your input
Your dispatch prompt contains:
plan_path— the path to the cluster's plan: arefactor_plan.mdfor a single group, or aconsolidated_plan.mdfor a merged cluster (under~/.ariadne/prioritize/<timestamp>/<fault_area>/refactor_plan.mdor~/.ariadne/prioritize/<timestamp>/clusters/<slug>/consolidated_plan.md).row_ids[]— thePlanTaskids for every row in the cluster (for a merged cluster, the union across its member groups). Every id must be claimed by exactly one authored task'splan_task_ids.output_path— where to write your assignment file, beside the plan (task_assignment.jsonin the same directory asplan_path).
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 4cb371c2f8c6
- 17d ago First seen · 186 lines · 169 tokens per session scan A f397a2bc3cf3
refactor-task-architect is an agent published in the GitHub repository CRJFisher/ariadne (23 stars, last pushed 4d ago), licensed MIT. It adds 169 tokens to every session and 2,359 once invoked, about $0.0007 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 2026-10-06.
Other agents, from other repositories
tech-lead
Tech Lead orchestrator — manages sprint dispatch, merges, and direct commits to main.
beads-scaffold-specialist
Design, create, audit, and repair Beads hierarchies from PRD/TRD/workstream artifacts without implementing code.
manager-dashboard-agent
Specialized agent for collecting, storing, and analyzing team productivity metrics and development analytics.
orchestrator
Demo worker-shaped orchestrator. Dispatches to implementer / reviewer / qa specialists per task class. Modeled on the real worker.md so that slice-1 q8rl regex (literal subagenttype) and slice-2 hw6j frontmatter (placeholder declaration) both have non-Claude-only fixture coverage.
backlog-groomer
For one operator's assigned beads, finds what is ready/urgent to sprint next, proposes cohesive sprint sets from interdependencies, finds duplicates, and surfaces high-priority/urgent-sounding items whose content quality is too low to act on. Full beads mutation authority (merge/close/defer/reject), always…
agent-architect
Maintains this repository's AI agents, skills, prompts, instructions, commands, hooks, and MCP configuration using Weld Agent Graph.