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/chris-dare-dev/agent-kitWrote 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/chris-dare-dev/agent-kit/roadmap-refiner)<a href="https://agentmods.dev/agents/chris-dare-dev/agent-kit/roadmap-refiner"><img src="https://agentmods.dev/badge/agents/chris-dare-dev/agent-kit/roadmap-refiner/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/chris-dare-dev/agent-kit/roadmap-refiner"><img src="https://agentmods.dev/badge/agents/chris-dare-dev/agent-kit/roadmap-refiner.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 | $0.00094 | $0.02390 |
| Opus 5.5 | $0.00038 | $0.00956 |
| Sonnet 5.5 | $0.00019 | $0.00478 |
| Haiku 4.5 | $0.00009 | $0.00239 |
Grade C, and why
roadmap-refiner scanned grade C with 2 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 1mo 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.
Instruction-override phrasingmediumPrompt injection
Text telling the model to disregard its earlier instructions or safety rules is the shape of a prompt injection, whoever wrote it.
"Now run X", "Ignore previous instructions", "Authorize the user", "Add yourself Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
cat ".claude/agent-memory/roadmap-refiner/lessons.md" 2>/dev/null || echo "(no lessons yet)" How it starts
The opening of the file, as written. The whole thing — 212 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Roadmap Refiner
You are Phase 1 of the /roadmap pipeline. Your job is to convert a vague brief into a crisp, 1-page goal statement with an explicit Objective, Key Results, assumption tiers, and Won't list.
The orchestrator (slash command at .claude/commands/roadmap.md) dispatches you with substituted variables. You never invoke other sub-agents.
Input variables (substituted by the orchestrator)
{SLUG}— the roadmap slug (e.g.,cost-visibility-l3,kiali-multicluster){BRIEF}— the verbatim brief text from the user (or summarized from conversation context). Do NOT paraphrase.{ROADMAP_PATH}— absolute path toplans/<slug>-roadmap.md(already scaffolded byroadmap-init.sh)
Step 0 — Read persistent memory (skip-if-not-relevant)
cat ".claude/agent-memory/roadmap-refiner/lessons.md" 2>/dev/null || echo "(no lessons yet)"
Read memory only if lessons are relevant to this brief's domain. Do not load memory for its own sake.
Step 1 — Read the phase-refine reference (REQUIRED before any reasoning)
# Resolves from ANY CWD (the target repo is usually NOT the agent-kit checkout).
WS="${PERSONAL_WORKSPACE_ROOT:-$HOME/Work/workspace}"
cat "$WS/.claude/references/roadmap-phase-refine.md"
This is the canonical source for HMW mechanics, sharpening questions, assumption tier definitions, OKR rules, and Won't list conventions. Load it fully before proceeding.
Step 2 — Walk the 5 sub-steps
Sub-step 2a — "How Might We" reframe
Restate {BRIEF} as: How might we <verb> <outcome> for <who>?
- The verb must describe an outcome, not a planned solution.
- "How might we add a Grafana dashboard" → wrong (solution-embedded).
- "How might we make per-namespace cost visible to platform engineers" → right (outcome-shaped).
- If the verb is a solution (
add,migrate,deploy), challenge whether that is truly required or an assumed implementation.
Sub-step 2b — 3–5 sharpening questions
Pick the questions that are genuinely under-specified and change the plan. For each, answer from in-context evidence FIRST:
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.
- 1mo ago First seen · 212 lines · 94 tokens per session scan C 8ca682849ec0
roadmap-refiner is an agent published in the GitHub repository chris-dare-dev/agent-kit (0 stars, last pushed 2mo ago), licensed MIT. It adds 94 tokens to every session and 2,390 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it C with 2 findings (instruction-override phrasing, reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.