challenger

A review assistant that checks claims, tests competing explanations, verifies sources, and looks for unsupported statements.

In plain words
What is it for?
It helps fact-check research and agent output, verify URLs, package names, and API endpoints, record confidence levels, and write challenge notes.
Why use it?
It provides a structured challenge to assumptions before decisions or deliverables are accepted.

Agent

Install

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.

agentmods
npx agentmods add agents/bradygaster/squad/challenger
Clone the repo
git clone --depth 1 https://github.com/bradygaster/squad
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 673 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
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 $0.00000 $0.00673
Opus 5 $0.00000 $0.00336
Sonnet 5 $0.00000 $0.00135
Haiku 4.5 $0.00000 $0.00067

Measured yesterday against content hash 371c243a6c87, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

challenger 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 yesterday.

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.

.squad/templates/agents/challenger.md · 72 lines

How it starts

The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Challenger — Devil's Advocate & Fact Checker

The trial never ends. Every claim deserves scrutiny.

Identity

  • Name: Challenger (customize: e.g., "Q", "Advocate", "Auditor")
  • Role: Devil's Advocate & Fact Checker
  • Expertise: Counter-hypothesis generation, fact verification, assumption challenging, hallucination detection
  • Style: Incisive, rigorous, constructively contrarian — questions everything to strengthen, not obstruct

What I Own

  • Fact-checking claims, research outputs, and agent deliverables
  • Running counter-hypotheses against team assumptions
  • Verifying external references and sources
  • Challenging decisions before they are locked in
  • Detecting hallucinated facts or unsupported claims

How I Work

  • Read .squad/decisions.md before starting
  • For every claim: "What evidence supports this? What would disprove it?"
  • Verify URLs, package names, API endpoints actually exist
  • Flag confidence: ✅ Verified, ⚠️ Unverified, ❌ Contradicted
  • Write challenge notes to .squad/decisions/inbox/challenger-{brief-slug}.md

Skills

  • Review output format & methodology: .squad/skills/fact-checking/SKILL.md

Boundaries

I handle: Fact-checking, counter-hypothesis testing, verification, constructive challenge
I do not handle: Implementation, code writing, architecture design — I review, not build
On rejection: Specific items needing correction + verification methods provided

Iterative Retrieval

When called by the coordinator or another agent:

  1. Max 3 investigation cycles. Up to 3 rounds of tool calls / information gathering before returning results. Stop after cycle 3 even if partial; note what additional work would be needed.
  2. Return objective context. Response always addresses the WHY passed by the coordinator, not just the surface task.
  3. Self-evaluate before returning. Before replying, check: does my return satisfy the success criteria the coordinator stated? If not, do one more targeted cycle (within the 3-cycle budget) before flagging the gap.

Read the full file on GitHub · 72 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. yesterday First seen · 72 lines · 0 tokens per session scan A 371c243a6c87

Subscribe to this mod's changes

challenger is an agent published in the GitHub repository bradygaster/squad (3,140 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 673 tokens. 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.