paper-chaser-mcp: Agent for Claude Code

.github/agents/mcp-stress-tester.agent.md

MCP Stress Tester is an agent for Claude Code from joshuasundance-swca/paper-chaser-mcp. It costs 69 tokens per session (2,881 once invoked), scanned A, original, from a forked repository, MIT.

A testing agent for checking how MCP research and discovery tools behave under difficult conditions. MCP is a standard way for agents to use external tools.

In plain words
What is it for?
It stress-tests research tools, reviews their input and output designs, measures extra data and agent steps, and reports usability problems to developers.
Why use it?
It helps find hallucinations, misleading status reports, oversized responses, and differences between a tool's stated and actual behavior.

Agent for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: model in frontmatter; mentions subagents.

This is joshuasundance-swca/paper-chaser-mcp's own configuration. It tells Claude Code how to work on paper-chaser-mcp itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything paper-chaser-mcp configures →

Reuse

Borrowing it

Nothing to install: this file belongs to joshuasundance-swca/paper-chaser-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/joshuasundance-swca/paper-chaser-mcp/master/.github/agents/mcp-stress-tester.agent.md
Clone the repo
git clone --depth 1 https://github.com/joshuasundance-swca/paper-chaser-mcp

Made for: Claude Code.

Wrote 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.

agentmods badge for MCP Stress Tester

README.md
[![agentmods](https://agentmods.dev/badge/agents/joshuasundance-swca/paper-chaser-mcp/mcp-stress-tester/github.svg)](https://agentmods.dev/agents/joshuasundance-swca/paper-chaser-mcp/mcp-stress-tester)
Your own site
<a href="https://agentmods.dev/agents/joshuasundance-swca/paper-chaser-mcp/mcp-stress-tester"><img src="https://agentmods.dev/badge/agents/joshuasundance-swca/paper-chaser-mcp/mcp-stress-tester/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.

agentmods 80×15 button for MCP Stress Tester

Your own site · 80×15
<a href="https://agentmods.dev/agents/joshuasundance-swca/paper-chaser-mcp/mcp-stress-tester"><img src="https://agentmods.dev/badge/agents/joshuasundance-swca/paper-chaser-mcp/mcp-stress-tester.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 69 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,881 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin fork From a forked repository.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00069 $0.02881
Opus 5 $0.00034 $0.01440
Sonnet 5 $0.00014 $0.00576
Haiku 4.5 $0.00007 $0.00288

Measured 12d ago against content hash 82f891bb4545, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

MCP Stress Tester 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 12d 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.

.github/agents/mcp-stress-tester.agent.md · 289 lines

How it starts

The opening of the file, as written. The whole thing — 289 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an expert AI systems tester and developer UX evaluator. Your job is to conduct relentless, multi-tiered stress tests of MCP research and discovery tools, uncover failure modes, quantify payload bloat, detect schema misdirection, identify transport/runtime divergence, and produce brutally honest developer-facing evaluations.

You are NOT a research assistant. You do not answer research questions. You TEST research tools and REPORT on their behavior to a developer audience.

Core Mandate

  1. Break things intentionally. Design queries that probe edges: data droughts, nonsensical intersections, pseudoscience, regulatory niche, hyper-specific extraction, and leading questions that invite hallucination.
  2. Measure everything. Track payload sizes, count file-read fallbacks, record redundant re-serialization, time how many agent turns each operation costs.
  3. Trust nothing. Verify that status flags (succeeded, verified, answered, grounded, fullTextObserved) actually match the content. If fullTextObserved: true but the follow-up abstains, that is schema misdirection — report it.
  4. Fail gracefully. When a tool abstains or returns partial results, evaluate whether the failure mode is clean (explicit metadata flags, clear unsupportedAsks) or dirty (silent data gaps, misleading confidence scores).
  5. Separate product defects from invocation defects. If the same guided contract behaves differently across the MCP client, the repo's native server entry point, and any local harness, treat that as a primary finding. Do not collapse transport/runtime divergence into a generic "tool quality" complaint.
  6. Do not let operational noise hide logic defects. Rate limits, auth gaps, hidden tools, and provider suppression can contaminate results. When they do, record them explicitly and distinguish them from ranking, synthesis, or schema problems.

Standard Test Protocol

Execute autonomously without asking for permission between phases.

Read the full file on GitHub · 289 lines

Changes

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.

  1. 12d ago First seen · 289 lines · 69 tokens per session scan A 82f891bb4545

Subscribe to this mod's changes

MCP Stress Tester is an agent published in the GitHub repository joshuasundance-swca/paper-chaser-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 69 tokens to every session and 2,881 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. It comes from a forked repository.

Related

Other agents, from other repositories

tdd-guide

Test-driven development guide for writing tests first, implementing the smallest passing change, and keeping verification tight. Use when the user explicitly wants TDD or when a task should be driven by failing tests before code.

Galaxy-Dawn/claude-scholar · 45 tokens

methodology-reviewer-agent

Serves as Peer Reviewer 1, a research methodology expert focusing on the rigor of research design: whether the methods answer the questions posed, the data collection and analysis are appropriate, the conclusions are supported, and the procedures are reproducible.

AlterLab-IEU/AlterLab-Academic-Skills · 0 tokens

code-reviewer

Performs iterative QA review of executed scripts. Verifies code correctness, methodology alignment, validation robustness, and output data quality. Creates parallel QA inspection scripts. Invoked by orchestrator after each Stage 5-8 script execution. Also performs QA review of profiling scripts during Data Onboarding…

brycewang-stanford/Auto-Empirical-Research-Skills · 70 tokens

verifier

End-to-end verification agent. Checks that slides compile, render, deploy, and display correctly. Use proactively before committing or creating PRs.

brycewang-stanford/Auto-Empirical-Research-Skills · 31 tokens

kingdee-qa-engineer

QA & Test Engineer for the kingdee-mcp project. Authors evals/ and tests/ cases, reproduces bugs against the live K3Cloud environment, and runs regression scans via bin/kmcp test.

WaHaiLong/KingdeeMCP · 49 tokens

tests

Test strategy, generation, authoring, and repair across all layers.

namastexlabs/automagik-tools · 15 tokens