grounded-research
Epistemic honesty protocol for code research. Forces uncertainty admission, cascading verification through authoritative sources (deepwiki, submodules, web), and approach validation before implementation.
npx skills add majiayu000/claude-skill-registry --skill grounded-research --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
# Grounded Research Protocol **Purpose**: Enforce epistemic honesty in code research. Admit uncertainty, cascade through authoritative sources, verify approaches BEFORE implementing. ## Core Doctrine: Uncertainty Admission > **You do not know what you have not verified.** ### Knowledge Cutoff Awareness | Fact | Implication | |------|-------------| | Today's date | 2026-01-16 | | Knowledge cutoff | May 2025 | | Gap | ~8 months | **For any library/API/pattern that could have changed since May 2025, you MUST:** 1. Admit uncertainty explicitly 2. Research via authoritative sources 3. Verify before claiming knowledge ### Uncertainty Markers Use these markers in your thinking and responses: | Marker | Meaning | Action | |--------|---------|--------| | `[UNCERTAIN]` | I'm not confident about this | Research required | | `[CUTOFF-GAP]` | This may have changed since May 2025 | Verify via sources | | `[VERIFIED]` | Confirmed via authoritative source | Safe to assert | | `[INFERRED]` | Based on patterns, not verification | State explicitly | ### Example Uncertainty Admission ``` [UNCERTAIN] I believe Effect.Service<>() uses double-parenthesis syntax, but my knowledge is from May 2025. Let me
- Core Doctrine: Uncertainty Admission
- Knowledge Cutoff Awareness
- Uncertainty Markers
- Example Uncertainty Admission
- Research Cascade Protocol
- When to Use Each Level
- Verification Query Patterns
- DON'T: Ask for "the answer"
- DO: Ask WHETHER your understanding is correct
- Verification Query Templates
- MCP Tool Usage
- deepwiki Tools
- Canonical Repos for deepwiki
- Submodule Research
Effect website (human-authored docs - HIGHEST trust) ls ../../submodules/website/content/src/content/docs/docs/ Effect source + tests (canonical patterns) find ../../submodules/effect/packages -name "*.test.ts" | head -20 effect-atom tests (atom patterns) ls ../../submodules/effect-atom/packages/atom/test/ Search for specific pattern grep -r "Effect.Service" ../../submodules/effect/packages/effect/src/
What does the grounded-research skill do?
Epistemic honesty protocol for code research. Forces uncertainty admission, cascading verification through authoritative sources (deepwiki, submodules, web), and approach validation before implementation.
How do I install it?
Run `npx skills add majiayu000/claude-skill-registry --skill grounded-research --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From majiayu000/claude-skill-registry, a repository with 534 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.