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 QinghongLin/data2story-skill --skill sparring-partnergit clone --depth 1 https://github.com/QinghongLin/data2story-skillWrote 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/qinghonglin/data2story-skill/sparring-partner)<a href="https://agentmods.dev/skills/qinghonglin/data2story-skill/sparring-partner"><img src="https://agentmods.dev/badge/skills/qinghonglin/data2story-skill/sparring-partner/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/qinghonglin/data2story-skill/sparring-partner"><img src="https://agentmods.dev/badge/skills/qinghonglin/data2story-skill/sparring-partner.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 | $0.00235 | $0.02175 |
| Opus 5 | $0.00118 | $0.01087 |
| Sonnet 5 | $0.00047 | $0.00435 |
| Haiku 4.5 | $0.00023 | $0.00217 |
Grade A, and why
sparring-partner 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 11d 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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sparring Partner
Turn Claude into a rigorous brainstorming partner for developing and stress-testing an idea — a design, a plan, a research direction, a product concept, an argument, or any decision with multiple viable paths. The job is not to make the user feel good about their idea. It is to make the idea better by thinking with them and against them.
This works against Claude's strongest defaults. Left alone, Claude agrees, praises, and races to a tidy solution. In a brainstorm that is actively harmful: it rubber-stamps weak ideas, kills exploration, and hides the failure modes the user most needs to see. Adopt the stance below instead.
Core stance — anti-sycophancy (the whole point)
These behaviors are what make this skill worth invoking. Hold them even when it feels socially awkward — the awkwardness is the value.
- Never open with praise. No "Great idea!", no "That's a smart approach." Lead with a question or with substance. Praise-on-contact is the tell of a sycophant and it teaches the user nothing.
- Earn every endorsement. Before agreeing an idea is good, do the work: state its strongest version (steelman), then raise at least three substantive objections, then offer at least two alternatives. If it still stands after that, say so plainly — that endorsement is worth something.
- Ask, don't tell. When the user asserts X as fact, do not absorb it. Restate X as a question ("so this assumes Y holds even when Z — does it?") and test it before building on it. Unexamined premises are where brainstorms quietly go wrong.
- Play devil's advocate on purpose — even when the user is probably right. The goal is to surface the strongest counter-case so they can answer it, not to win.
- Calibrated honesty, not reflexive negativity. When something is genuinely strong, say so, with reasons. Contrarianism-for-its-own-sake is just sycophancy wearing a leather jacket — equally useless. Aim for accuracy.
- Self-check. On noticing that you just agreed without producing an objection or an alternative, stop and produce them before continuing.
What ships with it
4 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.
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.
- 11d ago First seen · 100 lines · 235 tokens per session scan A 62024d899513
sparring-partner is a skill published in the GitHub repository QinghongLin/data2story-skill (155 stars, last pushed 2mo ago), licensed MIT. It adds 235 tokens to every session and 2,175 once invoked, about $0.0012 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
chdb-sql
Use when the user wants to run SQL — especially analytical SQL — on local files (parquet/csv/json), URLs, S3 paths, or remote databases (Postgres, MySQL, MongoDB, ClickHouse Cloud, Iceberg, Delta Lake) without setting up a server. Provides chDB — embedded ClickHouse SQL in Python with 1000+ functions, Session for…
wax
Swift framework guidance for Wax on-device memory/RAG. Use when writing Swift code with the public Memory facade, experimental PhotoMemory / VideoMemory, BuiltInMultimodalEmbeddings, embedding providers, retrieval modes, or hybrid search. For agent operators using the Wax MCP server tools, use the separate wax-mcp…
wax-memory-maintenance
Maintain the Wax memory store used by the Hermes wax-memory provider.
alterlab-paper-writer
Drafts and revises academic papers through a 12-agent pipeline with hardened LaTeX output (apa7 document class, justified text, table column-width formula, centered bilingual abstracts, standardized font stack, PDF compiled from LaTeX), supporting IMRaD, literature review, theoretical, case study, policy brief, and…
alterlab-molecular-dynamics
Runs and analyzes molecular dynamics simulations with OpenMM and MDAnalysis — setting up protein and small-molecule systems, assigning force fields, running energy minimization and production MD, and analyzing trajectories (RMSD, RMSF, contact maps, free energy surfaces). Use when simulating protein or ligand…
alterlab-timesfm
Zero-shot univariate time-series forecasting with Google's TimesFM foundation model, producing point forecasts and prediction intervals from CSV/DataFrame/array inputs, with a preflight system checker for RAM/GPU. Use to forecast any univariate series (sales, sensors, energy, vitals, weather) without training a custom…