quaere-evidence

quaere-evidence is a skill for Claude Code, Codex from haru0416-dev/quaere. It costs 98 tokens per session (2,385 once invoked), scanned A, original, MIT.

An evidence-first method for investigating bugs, failed builds, risky code reviews, and other uncertain engineering claims. It requires each conclusion to state the evidence, what could disprove it, and which check was run.

In plain words
What is it for?
Use it to investigate unclear bugs, flaky tests, CI failures, security or database changes, concurrency issues, external API changes, and large refactors.
Why use it?
It prevents patches and review decisions from being based only on guesses. The recorded checks make it easier to tell confirmed findings from untested or inconclusive ones.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Claude Code; mentions Codex; mentions OpenCode.

Good fit Use it to investigate unclear bugs, flaky tests, CI failures, security or database changes, concurrency issues, external API changes, and large refactors.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/haru0416-dev/quaere/quaere-evidence
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.

Any agent
npx skills add haru0416-dev/quaere --skill quaere-evidence
Clone the repo
git clone --depth 1 https://github.com/haru0416-dev/quaere

Made for: Claude Code, Codex.

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 quaere-evidence

README.md
[![agentmods](https://agentmods.dev/badge/skills/haru0416-dev/quaere/quaere-evidence/github.svg)](https://agentmods.dev/skills/haru0416-dev/quaere/quaere-evidence)
Your own site
<a href="https://agentmods.dev/skills/haru0416-dev/quaere/quaere-evidence"><img src="https://agentmods.dev/badge/skills/haru0416-dev/quaere/quaere-evidence/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 quaere-evidence

Your own site · 80×15
<a href="https://agentmods.dev/skills/haru0416-dev/quaere/quaere-evidence"><img src="https://agentmods.dev/badge/skills/haru0416-dev/quaere/quaere-evidence.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 98 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,385 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 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 5.1 $0.00098 $0.02385
Opus 5 $0.00049 $0.01192
Sonnet 5 $0.00020 $0.00477
Haiku 4.5 $0.00010 $0.00238

Measured 10d ago against content hash 44b1a857f405, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

quaere-evidence 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 10d 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.

skills/core/quaere-evidence/SKILL.md · 182 lines

How it starts

The opening of the file, as written. The whole thing — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Evidence-Gated Review

Iron Law

No patch, review comment, or root-cause conclusion without a falsifiable claim and one executed attempt to defeat it.

Output contract

Every claim or hypothesis acted on carries the four lines below. Everything else about format is the investigator's choice.

Lightweight evidence pass

Claim: <what you assert>
Evidence: <file:line / log / repro / trace>
Falsifier: <the observation that would defeat this>
Probe: <the check you ran trying to defeat it → its result>

Decisions

  • confirmed — the probe ran and did not defeat the claim.
  • confirmed (rebuttal-substituted) — the probe is unsafe or unavailable in scope (production replay, payments, destructive ops); the strongest counter-explanation was argued from source, tests, or spec, and failed. Label it exactly so the substitution is auditable.
  • rejected / inconclusive / deferred — when the Probe line says "not run", these are the only legal labels.

When to use

  • CI failures, flaky tests, regressions, or bugs whose cause is not already proven.
  • Risky PR reviews, security claims, API contract concerns, database/concurrency changes, multi-file refactors.
  • Any claim that would create noisy review feedback or a misleading handoff if published unverified.

When NOT to use

  • Typos, formatting-only edits, or a tiny user-requested edit where cause and fix are already known.
  • Pure code comprehension with no claim to decide — use quaere-semantic.
  • Version-sensitive external facts whose truth depends on current docs or SDKs — hand that sub-claim to quaere-grounding, then resume.

Probe budget

One discriminating probe per decision — the check whose outcome differs between the claim and its best alternative. Two at most; if both are inconclusive, stop and hand off. Supporting evidence never substitutes for the disconfirming attempt. Running a flaky test N times for a pass rate is one probe with a sample size, not N retries.

Read the full file on GitHub · 182 lines

Files

What ships with it

10 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.

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. 10d ago First seen · 182 lines · 98 tokens per session scan A 44b1a857f405

Subscribe to this mod's changes

quaere-evidence is a skill published in the GitHub repository haru0416-dev/quaere (5 stars, last pushed 1mo ago), licensed MIT. It adds 98 tokens to every session and 2,385 once invoked, about $0.0005 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 2026-08-31.

Related

Other skills, from other repositories

checking-code-quality

Checks code quality metrics including complexity, duplication, naming conventions, and function length. Use when running quality gates, reviewing code smells, or checking lint rules. Automatically triggered on complex modules or post-refactor.

telagod/code-abyss · 45 tokens

indexing-code

Code relationship graph and temporal intelligence via abyss CLI. Provides caller tracing, impact analysis, hotspot detection, and file-level context gathering. Agent automatically runs abyss before modifying code to check impact. Works with any agent that has shell access.

telagod/code-abyss · 50 tokens

patch-edit

Apply one rewrite-plan step at a time as small reviewable file edits, never rewriting whole files when a localised change suffices.

nexu-io/open-design · 30 tokens

hunk-extensions

Maps the hunkdiff/extension authoring surface for Hunk, the terminal diff viewer — hiding or reordering reviewed files, docked panes, alternate file views, commands and key bindings, dialogs, workspace writes, themes, syntax languages, VCS backends, lifecycle events. Use when writing, debugging, or installing a Hunk…

modem-dev/hunk · 105 tokens

printing-press-output-review

Internal sub-skill: agentic review of a printed CLI's sampled command output for plausibility issues that rule-based checks can't encode (substring-match relevance, format bugs, silent source drops, ranking failures). Invoked via the Skill tool by the main printing-press skill at Phase 4.85 and printing-press-polish…

mvanhorn/cli-printing-press · 102 tokens

fix

Dispatch fix subagent for FIX-FIRST gaps from review, re-review, then diagnose unresolved failures after 2 loops.

automagik-dev/genie · 25 tokens