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 skills add YoungjaeDev/my-claude-plugins --skill research-orchestratorgit clone --depth 1 https://github.com/YoungjaeDev/my-claude-pluginsWrote 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/skills/youngjaedev/my-claude-plugins/research-orchestrator)<a href="https://agentmods.dev/skills/youngjaedev/my-claude-plugins/research-orchestrator"><img src="https://agentmods.dev/badge/skills/youngjaedev/my-claude-plugins/research-orchestrator/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/skills/youngjaedev/my-claude-plugins/research-orchestrator"><img src="https://agentmods.dev/badge/skills/youngjaedev/my-claude-plugins/research-orchestrator.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 | $0.00145 | $0.04386 |
| Opus 5 | $0.00072 | $0.02193 |
| Sonnet 5 | $0.00029 | $0.00877 |
| Haiku 4.5 | $0.00015 | $0.00439 |
Grade A, and why
research-orchestrator scanned grade A with 1 finding 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 4d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
When no delegation channel is available, run each chosen axis yourself, one at a time, following its `axis-contracts.md` contract with your own tools (`gh`, `curl` / `uvx hf`, exa / WebSearch, Context7 / DeepWiki, paper- How it starts
The opening of the file, as written. The whole thing — 248 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research Orchestrator
Entry point for scout v2.0. Replaces the legacy scout / deep-scout agents with a small router that fans out to specialized scouts and consolidates via synthesis-scout.
When to use vs. skip
Use when:
- The query needs more than one source axis (GitHub + docs, web + HF, etc.)
- The user explicitly asks for a research report or comparison
- The user says "thorough", "deep", "comprehensive", "compare", "best practices", "research"
- A single-axis quick lookup is fine: the orchestrator will detect that and fan out to just one scout
Skip / use something else when:
- The user wants a GitHub PR review →
dev:cr-fix/dev:resolve-issue - The user wants to ask a single question about one repo →
scout:askdirectly - The user wants to read library API docs only →
context7MCP directly - The user wants a single paper's PDF or full text →
scout:paper-searchdownload_*/read_*directly (paper-scout is metadata-only) - The user's query is outside the code / ML domain (politics, market, history, biographies, general policy) →
/deep-researchdirectly. Its 7-phase + adversarial verify + state-machine flow is tuned for generic topics; scout's 5-axis routing is tuned for code/ML and would mis-route on those queries. Orchestrator does not delegate to/deep-research: the boundary is intentional.
See references/agent-routing.md for the full should / should-NOT matrix.
Inputs
query(required): user's natural-language research targetmode(optional):quick|deep; auto-detected if absentworkspace_dir(optional): when set, the orchestrator reuses this directory and the caller owns lifecycle (enables partial re-execution; see Phase 0 below). When unset, each run gets its ownmktempdirectory.report_path(optional): defaults to${workspace_dir}/final_report.md
Data-transfer protocol: file-based via $WORKSPACE/{NN}_{axis}.json. Sibling scouts do not communicate directly; everything flows through the workspace and is merged by synthesis-scout. Lexical filename ordering = deterministic merge order.
What ships with it
13 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/agent-routing.md 6.5 KB
- references/axis-contracts.md 8.6 KB
- references/brightdata-cli-commands.md 7.4 KB
- references/brightdata-guide.md 22 KB
- references/brightdata-mcp-setup.md 4.5 KB
- references/brightdata-mcp-tools.md 14 KB
- references/exa-web-search.md 8.3 KB
- references/github_api.md 3.0 KB
- references/huggingface_api.md 4.5 KB
- references/resource-finder.md 6.2 KB
- references/synthesis-rules.md 5.0 KB
- scripts/search_github.py 7.0 KB runs code
- scripts/search_huggingface.py 9.1 KB runs code
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.
- 4d ago Changed · -12 lines · -93 tokens per session de89cea6588a
- 9d ago First seen · 260 lines · 238 tokens per session scan A c62010007a64
research-orchestrator is a skill published in the GitHub repository YoungjaeDev/my-claude-plugins (2 stars, last pushed 2d ago), licensed MIT. It adds 145 tokens to every session and 4,386 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…