← All creators

Ahmad-Jaradat-Space

User

Evidence-first research skills for Claude Code and Codex. Keeps agent claims tied to a verifiable evidence graph instead of chat history.

3 indexed · 0 Featured · 0 stars · avg score 57

Categories

Indexed Skills (3)

AI & Automation Listed

evidence-first-claims

Record sources, evidence, hypotheses, results and claims in a ScientistOS research graph at the E0-E5 evidence level the lineage actually supports, and split work between the Scientist and Auditor roles. Use before asserting any scientific finding, citing a paper, or saying something "shows" or "causes" or "improves" something, and when promoting a claim, writing an abstract or results section, or auditing someone else's conclusion. Triggers include "record this finding", "add this evidence", "is this claim supported", "can we say X causes Y", "write up the results", "what level is this claim", "check for confounding or leakage". Do not use for non-scientific assertions or projects without a ScientistOS research graph.

0 Updated 2 months ago
Ahmad-Jaradat-Space
AI & Automation Listed

research-graph

Drive a ScientistOS research graph (research.db) using the `science` CLI and the scientistos MCP tools. Use when work touches a research question, literature or paper search, a scientific corpus, an experiment, a result, or a manuscript, and before writing any scientific conclusion so it has graph lineage rather than only chat history. Triggers include "start a research question", "search the literature", "find papers on", "what does the literature say", "index this paper/PDF", "what depends on this claim", "show the research status", "export the graph", "package this run for replay". Do not use for ordinary software tasks, general web research, or any project without a ScientistOS research graph.

0 Updated 2 months ago
Ahmad-Jaradat-Space
AI & Automation Listed

research-integrity-gate

Diagnose and repair a failing ScientistOS integrity gate, provenance chain, or replay verification. Use when `science integrity`, `verify-store` or `verify-replay` fails, when the Claude Code Stop hook or the git pre-commit hook blocks on a broken research graph, before handing research to anyone, and before treating a result as final. Triggers include "integrity check failed", "the hook is blocking me", "blocking finding", "graph digest", "provenance chain broken", "claim has no lineage", "evidence not verified", "verify the replay package", "is this research ready to share". It only ever repairs graphs; never use it to weaken or bypass a check. Not for general CI, lint or test failures.

0 Updated 2 months ago
Ahmad-Jaradat-Space

Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.