sense

sense is a skill for Claude Code, Codex from m3data/sense-mcp. It costs 21 tokens per session (375 once invoked), scanned A, original, Apache-2.0.

A semantic search tool for finding related information in indexed project records, documentation, research, and code.

In plain words
What is it for?
Use it to investigate how something works, find architecture decisions, locate deployment documentation, or trace related code and research.
Why use it?
It helps when you know the topic but not the exact words used in the stored material. Filters can narrow results to a project or source type.

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/m3data/sense-mcp/sense
Any agent
npx skills add m3data/sense-mcp --skill sense
Clone the repo
git clone --depth 1 https://github.com/m3data/sense-mcp

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for sense

README.md
[![agentmods](https://agentmods.dev/badge/skills/m3data/sense-mcp/sense.svg)](https://agentmods.dev/skills/m3data/sense-mcp/sense)
Your own site
<a href="https://agentmods.dev/skills/m3data/sense-mcp/sense"><img src="https://agentmods.dev/badge/skills/m3data/sense-mcp/sense.svg" alt="Measured on agentmods" height="20"></a>
Per session 21 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 375 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 $0.00021 $0.00375
Opus 5 $0.00010 $0.00187
Sonnet 5 $0.00004 $0.00075
Haiku 4.5 $0.00002 $0.00038

Measured 4d ago against content hash c18091f2d681, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

sense 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 4d 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.

skills/sense/SKILL.md · 40 lines

What it actually says

Search the indexed ecosystem using the sense_search MCP tool.

With arguments

Parse $ARGUMENTS as follows:

  • The main text is the search query
  • --project <name> filters to a specific project
  • --type <source_type> filters by type (trace, documentation, project_claude, reference, research, teaching, code)
  • --limit <n> sets max results (default 10)
  • --mode <mode> overrides Vibe Harness mode for search shaping (explore, build, think-with, ship, cool-off, none). Omit to auto-detect. Use none to force flat cosine search regardless of Vibe Harness state.

Examples:

  • /sense how does authentication work
  • /sense API error handling --project backend
  • /sense deployment configuration --type documentation --limit 5
  • /sense architecture decisions --mode explore

Without arguments

When $ARGUMENTS is empty, synthesize a search query from the conversation context:

  1. Review the last 3-5 exchanges in the current conversation
  2. Identify the core topic, concepts, or question being discussed
  3. Compose a 3-8 word query that captures the semantic thread
  4. Prefix results with "Searching for: [synthesized query]" so the user sees what was searched
  5. Call sense_search with the synthesized query

This enables mid-conversation discovery — just type /sense to surface related prior work without crafting a query.

General

Call sense_search with the parsed arguments and display the results. If the index is empty, suggest running /sense-sync first.

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. 4d ago First seen · 40 lines · 21 tokens per session scan A c18091f2d681

Subscribe to this mod's changes

sense is a skill published in the GitHub repository m3data/sense-mcp (1 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 21 tokens to every session and 375 once invoked, about $0.0001 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 2026-08-31.