evidence-based-reasoning
Enforces 100% factual, evidence-backed reasoning — no inference or paraphrase, all claims must be supported by quoted source code with file paths and line…
Skill by MoltenRuby
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/moltenruby-evidence-based-reasoning
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: ml
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/evidence-based-reasoning
What's inside
- Skill name
evidence-based-reasoning- Description
- Enforces 100% factual, evidence-backed reasoning — no inference or paraphrase, all claims must be supported by quoted source code with file paths and line numbers, or external URLs with confirmed version matches
- License
- MIT
- Compatibility
- opencode
- Instructions
- 29 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
## Evidence-based reasoning discipline
- **100% factual and verbatim.** Never infer, guess, or paraphrase what code does. Quote file paths, function names, variable names, and logic exactly as they appear in the source. If you have not read the relevant source file, say so and read it before drawing any conclusion.
- **No speculation.** Every claim about system behaviour must be backed by at least one of:
- A quoted file path and line number from the actual source
- A captured program output recorded verbatim
- An external URL with an exact version match to the project in use
### Version-confirmed external references
Before citing any public issue, forum post, or changelog entry as evidence:
1. Verify the exact version of the library or runtime in use (read `package.json`, `go.mod`, `Cargo.toml`, `requirements.txt`, `pyproject.toml`, `pom.xml`, or equivalent).
2. The version in the project must match or overlap the affected range stated in the external report.
3. If the version cannot be confirmed, state this explicitly and do not treat the external reference as confirmed evidence.
### Citing sources
- Source code: `<file path>:<line number> — <verbatim quoted content>`
- Program output: record verbatim, label as `EV-{id}`
- External reference: `<URL> (version range: <x.y.z – x.y.z>)`
### Unknown information
If the relevant source has not been read, or the relevant output has not been captured, state this explicitly:
> "I have not yet read `<file>`. I will read it before drawing a conclusion."
Never fill this gap with inference.From MoltenRuby/MrSmith, MIT.
More from MoltenRuby/MrSmith
- 0.414score
copilot-search
MoltenRuby
Use when you need to find code or documentation by concept or meaning and don't know which file contains it.
SkillDFreshness: DClaude Code +9 - 0.406score
phase-test-gate
MoltenRuby
Defines the mandatory test-gate convention for phased execution plans — every phase must end with a [TEST] step that acts as a hard gate before the next phase…
SkillDFreshness: DClaude Code +9 - 0.406score
structured-labels
MoltenRuby
Defines the standard structured label conventions used across all analytical, planning, and debugging agents — REQ, RISK, BS, HYP, EV identifiers with…
SkillDFreshness: DClaude Code +9 - 0.406score
input-mode-detection
MoltenRuby
Defines four input modes (analysis-only, plan-only, both, neither) and the corresponding behaviour for each — used by programmer agents to determine how to…
SkillDFreshness: DClaude Code +9 - 0.406score
nomad-jobs
MoltenRuby
Author new Nomad v1.11.2 HCL job specs, review and validate existing specs, and convert Docker Compose files to Nomad HCL — service, batch, and periodic/cron…
SkillDFreshness: DClaude Code +9 - 0.406score
programmer-behaviour-rules
MoltenRuby
Standard pre-coding behaviour rules for programmer agents — read before writing, match conventions, no new dependencies without reason, explain before…
SkillDFreshness: DClaude Code +9 - 0.406score
opencode-artefact-conventions
MoltenRuby
Stable OpenCode configuration conventions for agents, skills, commands, and rules — frontmatter fields, file placement, naming constraints, and the…
SkillDFreshness: DClaude Code +9 - 0.406score
consensus-reviewer-output-format
MoltenRuby
Standard output format for consensus reviewer agents — Verdict/Reasons/Suggested changes structure with AGREE/DISAGREE verdicts
SkillDFreshness: DClaude Code +9 - 0.406score
deviation-reporting
MoltenRuby
Rules for classifying and reporting deviations from a plan during implementation — breaking deviations require user approval, trivial deviations are…
SkillDFreshness: DClaude Code +9 - 0.406score
feature-doc-schemas
MoltenRuby
Canonical schemas for requirements.md, analysis.md, SOP.md, acceptance-criteria.md, and doc/index.md — exact structure, section headings, and field formats…
SkillDFreshness: DClaude Code +9 - 0.406score
git-commit-workflow
MoltenRuby
Standard Git workflow conventions for commits, branches, and pull requests — commit message format, branch naming, squash rules, and PR hygiene to follow when…
SkillDFreshness: DClaude Code +9 - 0.402score
communication-rules
MoltenRuby
Universal communication rules for analytical and workshop agents — one question at a time, no silent assumptions, structured responses, no filler language
SkillDFreshness: DClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Mar 21, 2026 (7 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/MoltenRuby/MrSmith
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from MoltenRuby
- 0.414score
copilot-search
MoltenRuby
Use when you need to find code or documentation by concept or meaning and don't know which file contains it.
SkillDFreshness: DClaude Code +9 - 0.406score
opencode-artefact-conventions
MoltenRuby
Stable OpenCode configuration conventions for agents, skills, commands, and rules — frontmatter fields, file placement, naming constraints, and the…
SkillDFreshness: DClaude Code +9 - 0.406score
phase-test-gate
MoltenRuby
Defines the mandatory test-gate convention for phased execution plans — every phase must end with a [TEST] step that acts as a hard gate before the next phase…
SkillDFreshness: DClaude Code +9 - 0.406score
deviation-reporting
MoltenRuby
Rules for classifying and reporting deviations from a plan during implementation — breaking deviations require user approval, trivial deviations are…
SkillDFreshness: DClaude Code +9 - 0.406score
consensus-reviewer-output-format
MoltenRuby
Standard output format for consensus reviewer agents — Verdict/Reasons/Suggested changes structure with AGREE/DISAGREE verdicts
SkillDFreshness: DClaude Code +9 - 0.406score
feature-doc-schemas
MoltenRuby
Canonical schemas for requirements.md, analysis.md, SOP.md, acceptance-criteria.md, and doc/index.md — exact structure, section headings, and field formats…
SkillDFreshness: DClaude Code +9
Cite this page
evidence-based-reasoning skill. Agent Mart. https://agentmart.dev/skills/moltenruby-evidence-based-reasoning (last updated 2026-10-24)