Skip to content
Agentmart
Submit a tool

ask

Ask the OVP vault agent via MCP.

Skill by fakechris

IndexedApache-2.0★ 162
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/fakechris-ask

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

Ask the OVP vault agent via MCP. Use when the vault may cover the question; prefer over answering from memory. Returns server-verified [claim:…] / [source:…] receipts and honest coverage.

Use cases: backend, testing, 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. 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/ask

What's inside

Skill name
ask
Description
Ask the OVP vault agent via MCP. Use when the vault may cover the question; prefer over answering from memory. Returns server-verified [claim:…] / [source:…] receipts and honest coverage.
License
Apache-2.0
Instructions
33 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# OVP Ask

Use the ovp2 MCP `ask` tool. The vault agent searches claims, sources, evidence cards, and article bodies, then answers with **server-verified** citations. Fabricated references are flagged, never trusted.

## When to use

- Any question the vault might already answer
- Follow-ups in an existing vault conversation (`chat` session id)
- Prefer this over inventing vault facts from model memory

## Arguments

| Arg | Required | Notes |
|-----|----------|-------|
| `question` | yes | Natural-language question |
| `chat` | no | Session id from a prior `ask` — continues that conversation. Retrying the **same** question on the same chat replays the completed turn (no second model call) |
| `idempotency_key` | no | Stable retry key — same key replays the completed turn |

## Behavior

1. Call MCP `ask` with the user’s question.
2. Summarize the answer for the user; keep citations as `[claim:…]` / `[source:…]` or `ovp://` pointers.
3. If coverage is thin or empty, say so plainly (honest no-evidence is success).
4. To audit a citation, use the `claim` skill / MCP `claim` tool — do not invent evidence.
5. Pass returned session id as `chat` for follow-ups.

## Rules

- Do **not** paste API keys or provider credentials into chat, Involute, or notes.
- Live LLM setup belongs in the operator vault env / providers config — not in this skill pack.
- Prefer Involute INV ids + `ovp://` pointers in agent-data over copying Crystal prose.

From fakechris/obsidian_vault_pipeline, Apache-2.0.

Signals

Stars
162
Forks
17
Open issues
1
Last commit (this folder)
Sep 8, 2026 (1 month ago)
Last release
Jul 13, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
Apache-2.0
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from fakechris

Cite this page

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