Skip to content
Agentmart
Submit a tool

epistemics

Calibrate historical or uncertain investigation findings, distinguish evidence from inference, and report source coverage and gaps.

Skill by draht-dev

IndexedMIT★ 7
Write a review

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/draht-dev-epistemics

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.

We use your email only for this newsletter. Privacy

About

Calibrate historical or uncertain investigation findings, distinguish evidence from inference, and report source coverage and gaps.

Use cases: documents, testing, ml, agent-building

No 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 and code. 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/epistemics

What's inside

Skill name
epistemics
Description
Calibrate historical or uncertain investigation findings, distinguish evidence from inference, and report source coverage and gaps.
License
MIT
Instructions
21 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Epistemics

Cite evidence for claims about history or intent. Code demonstrates behavior; motivation requires a decision record, commit/PR discussion, source comment, or other relevant evidence. When those are absent, label the inference.

Use these tiers when they help communicate uncertainty:

- **Direct:** a cited explicit statement from a source in a position to know.
- **Supported:** independent indirect sources converge; explain that support without presenting it as an explicit statement.
- **Inferred:** a plausible interpretation of evidence; state the inference and what it rests on.
- **Speculative:** an unresolved hypothesis, separate from established findings.
- **Unknown:** evidence was unavailable or searches did not resolve the question.

Treat the user's suggested explanation as a candidate. Seek evidence that could contradict the leading interpretation. Surface conflicting sources instead of choosing a tidier narrative. Repeated code can reflect convention or copying; it does not establish intent by itself.

Distinguish a search with no result from a source that was not searched. A null result means no evidence was found within the stated scope, not that no evidence exists. Report material gaps; do not invent a gap to satisfy a report template.

For an investigation report, include a concise **Sources Consulted** summary with relevant queries, coverage, null results, and unavailable sources. Scope source collection to the question; do not require every service or evidence category for every answer. Follow any caller-specified report format.

Use `verification-gate` for present-state claims that can be checked by execution. Preserve `verify-work`'s observed/derived/assumed labels and verdict-first ordering when that command is selected.

From draht-dev/draht, MIT.

More from draht-dev/draht

Signals

Stars
7
Forks
0
Open issues
2
Last commit (this folder)
Oct 21, 2026 (8 days ago)
Last release
Oct 24, 2026 (5 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from draht-dev

Cite this page

epistemics skill. Agent Mart. https://agentmart.dev/skills/draht-dev-epistemics (last updated 2026-10-24)