ask

A vault question-answering skill for quickly finding information in a collection of notes. It searches notes for decisions, blockers, documents, and project status.

In plain words
What is it for?
Use it for questions such as what was decided about authentication, who is blocked, which documents cover a topic, or the status of a project.
Why use it?
It avoids manually opening many files to find what was decided, what is blocked, or where a document is stored. It can use combined keyword and meaning-based search when available, with a simpler text search as a fallback.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/kv0906/pm-kit/ask
Any agent
npx skills add kv0906/pm-kit --skill ask
Clone the repo
git clone --depth 1 https://github.com/kv0906/pm-kit

Made for: Claude Code, Codex.

Per session 51 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 835 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00051 $0.00835
Opus 5 $0.00026 $0.00417
Sonnet 5 $0.00010 $0.00167
Haiku 4.5 $0.00005 $0.00084

Measured 2d ago against content hash 15b886c7080f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ask scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/ask/SKILL.md · 112 lines

How it starts

The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/ask — Question Answering

Fast answers from your vault. Uses QMD hybrid search (BM25 + vector + rerank) when available, with automatic fallback to vault grep.

Context

Config: @_core/config.yaml

Input

User input: $ARGUMENTS

Processing Steps

Step 1: Detect Search Backend

Check if QMD MCP tools are available in the current session:

  1. Try qmd_status (MCP tool) to check QMD availability and index state.
  2. Evaluate result:
    • QMD available AND has embedded docs → use QMD mode
    • QMD available but 0 embedded docs → use Fallback mode + hint
    • QMD not available (tool missing/error) → use Fallback mode

Step 2: Parse Question

  • Detect project if mentioned
  • Identify question type:
    • "what did we decide about X" → decisions
    • "who's blocked" → blockers
    • "find doc for X" → docs
    • "status of X" → index

Step 3A: QMD Mode (preferred)

When QMD is available and indexed:

  1. Deep search: Call qmd_deep_search with the user's question.
    • Use collection filter from config: _core/config.yaml → qmd.collection_name (default: pm-kit)
    • Cap results: use qmd.max_results from config (default: 8)
  2. Retrieve top docs: Call qmd_get or qmd_multi_get for the top-scored results.
    • Apply minimum score threshold from config: qmd.min_score (default: 0.35)
  3. Synthesize answer: Read the retrieved content and produce a grounded answer with source citations.

Step 3B: Fallback Mode (vault grep)

When QMD is not available or not indexed:

  1. Search Strategy

    Question Type Search Path
    Decisions decisions/{project}/*.md
    Blockers blockers/{project}/*.md
    Docs docs/{project}/*.md, docs/general/*.md
    Status 01-index/{project}.md
    General All folders
  2. Search Methods

    • Filename match first (fastest — naming-as-API)
    • Frontmatter field match
    • Content grep (slower)

Step 4: Return Answer

Read the full file on GitHub · 112 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 112 lines · 51 tokens per session scan A 15b886c7080f

Subscribe to this mod's changes

ask is a skill published in the GitHub repository kv0906/pm-kit (133 stars, last pushed 2mo ago), licensed MIT. It adds 51 tokens to every session and 835 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

critique-agent

Pressure-test an existing product brief or PRD pack — find gaps, hidden assumptions, inconsistencies, and failure modes before stakeholder review. Use when: critique brief, critique PRD, devils advocate, red team, pressure test, find holes, what could go wrong, stress test the doc, pre-mortem.

ertra/pm-brain-for-cursor · 69 tokens

interview-frameworks

Frameworks for user interviews, question design, and qualitative research. Use when conducting user interviews, designing interview guides, researching user needs, or gathering qualitative insights. Trigger on: 'create an interview guide', 'how do I interview users', 'customer discovery questions', 'user research…

ertra/pm-brain-for-cursor · 69 tokens

query-datasets

Answer grounded yes/no questions about what exists in the 03-datasets/ example corpora (supporttickets, calltranscripts) using the local SQLite + FTS5 + vector hybrid index. Use when the user asks 'do/does [corpus] contain/mention/include/have requests for X?', 'any [corpus] about Y?', or similar factual queries over…

ertra/pm-brain-for-cursor · 91 tokens

critique-prd

Rubric-score a PRD pack (02 + 03) against the 7-dimension PM Brain rubric and run a 4-persona panel review. Returns scores, panel critiques, the single weakest section, a concrete rewrite, and P0/P1 fix lists. Use when: PRD review, score my PRD, rubric review, panel review, rewrite weakest section, critique-prd.

ertra/pm-brain-for-cursor · 86 tokens

create-prd

Write the PRD pack (02-product-requirements.md + 03-success-metrics.md) on top of an approved product brief. Use when: PRD, product requirements, requirements doc, success metrics, spec writeup, shape the PRD.

ertra/pm-brain-for-cursor · 56 tokens

create-internal-feature-announcement

Drafts an internal feature announcement (IFA) from the PRD pack and user sources using the repo template, then writes 04-internal-feature-announcement.md. Use when: IFA, internal feature announcement, internal FAQ, Slack IFA prep, launch comms pack, or internal product documentation from the template.

ertra/pm-brain-for-cursor · 71 tokens