q-challenge

A command that challenges the assumptions and explanations behind an investigation by comparing them with supporting and opposing evidence.

In plain words
What is it for?
Use it during an investigation to review hypotheses, inspect findings and evidence, classify claims as facts or assumptions, and identify tests that could disprove them.
Why use it?
It helps catch confirmation bias, circular reasoning, mistaken identity matches, and conclusions that have not been tested against disproof.

Command

Part of the huntkit plugin — 2 skills, 22 commands, 2 hooks shipped together

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 commands/assafkip/huntkit/q-challenge
Clone the repo
git clone --depth 1 https://github.com/assafkip/huntkit

Or install huntkit, the plugin that ships this one along with the rest of its 2 skills, 22 commands, 2 hooks.

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 398 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 $0.00398
Opus 5 $0.00000 $0.00199
Sonnet 5 $0.00000 $0.00080
Haiku 4.5 $0.00000 $0.00040

Measured 3d ago against content hash 8e648cf0d181, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

q-challenge 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 3d 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.

commands/q-challenge.md · 40 lines

What it actually says

Force a structured challenge of all investigation assumptions and hypotheses.

This is a mid-investigation health check. Do NOT skip steps.

Steps

  1. Read canonical/scope.md for hypotheses
  2. Read ALL files in investigations/<case>/findings/
  3. Read investigations/<case>/evidence/ index (list files, read key ones)
  4. Read memory/investigation-state.md for current assessment

For each hypothesis, answer:

Question Answer
What evidence SUPPORTS this? (cite specific files)
What evidence CONTRADICTS this? (cite specific files)
What would DISPROVE this? (describe the test)
Have we actually looked for disproving evidence? Yes/No
Is this based on fact, inference, or assumption? (classify)
Could the same evidence support a different explanation? (describe)

Then check for:

  • Circular reasoning: Are we using conclusion A to support conclusion B, and B to support A?
  • Name-matching traps: Are we assuming two accounts/profiles are the same person just because the name matches?
  • Confirmation bias: Have we only collected evidence that supports our theory?
  • Source independence: Do multiple "sources" actually trace back to the same original data point?
  • Community inference vs. identification: Did witnesses IDENTIFY someone, or INFER from context?

Output

Write to investigations/<case>/findings/CHALLENGE-YYYY-MM-DD.md with:

  • Each hypothesis: current confidence, revised confidence (if changed), and why
  • Top 3 weakest assumptions that need testing
  • Recommended collection to test (not confirm) the weakest assumptions
  • Tag each recommended action with Energy + Time Est

Present the challenge to the founder as choices, not commands.

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. 3d ago First seen · 40 lines · 0 tokens per session scan A 8e648cf0d181

Subscribe to this mod's changes

q-challenge is a command published in the GitHub repository assafkip/huntkit (51 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 398 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.