tachi: Skill for Claude Code

.claude/skills/tachi-shared/SKILL.md

tachi-shared is a skill for Claude Code from davidmatousek/tachi. It costs 56 tokens per session (724 once invoked), scanned A, original, Apache-2.0.

A shared set of reference material for security-analysis agents. It defines threat categories, severity levels, and the standard format for recording security findings.

In plain words
What is it for?
Use it when building or running a security-analysis pipeline that classifies threats, rates their severity, or passes findings between agents.
Why use it?
It keeps different security checks consistent instead of letting each agent use its own definitions and report structure.

Skill for Claude Code

Written for Claude Code: installed under .claude/. Also seen: reads .claude/ paths.

This is davidmatousek/tachi's own configuration. It tells Claude Code how to work on tachi itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything tachi configures →

Reuse

Borrowing it

Nothing to install: this file belongs to davidmatousek/tachi. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/davidmatousek/tachi/main/.claude/skills/tachi-shared/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/davidmatousek/tachi

Made for: Claude Code.

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 tachi-shared

README.md
[![agentmods](https://agentmods.dev/badge/skills/davidmatousek/tachi/tachi-shared.svg)](https://agentmods.dev/skills/davidmatousek/tachi/tachi-shared)
Your own site
<a href="https://agentmods.dev/skills/davidmatousek/tachi/tachi-shared"><img src="https://agentmods.dev/badge/skills/davidmatousek/tachi/tachi-shared.svg" alt="Measured on agentmods" height="20"></a>
Per session 56 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 724 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 5.1 $0.00056 $0.00724
Opus 5 $0.00028 $0.00362
Sonnet 5 $0.00011 $0.00145
Haiku 4.5 $0.00006 $0.00072

Measured 3d ago against content hash 8d529d20137a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

tachi-shared 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 3d 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/tachi-shared/SKILL.md · 40 lines

How it starts

The opening of the file, as written. The whole thing — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Tachi Shared References

Cross-cutting domain knowledge consumed by multiple tachi agents. These reference files are the canonical source of truth for definitions that appear across the pipeline -- severity bands, STRIDE+AI threat categories, and finding format conventions. Individual agents load specific references on-demand using the Read tool rather than duplicating definitions inline.

Domain Coverage

This skill contains four shared reference files:

  1. Severity Bands -- Composite score thresholds, severity band boundaries, color codes, SLA mappings, disposition defaults, and review date calculations. Used everywhere severity classification drives decisions.

  2. STRIDE+AI Categories -- All 11 threat categories (6 STRIDE + 5 AI) with descriptions, ID prefixes, OWASP references, and the DFD element-to-category applicability matrix that determines which categories apply to which component types.

  3. Finding Format -- The finding intermediate representation (IR) specification: required fields, optional fields, ID format conventions, validation rules, and table output format. Defines the contract between threat agents (producers) and downstream consumers.

  4. MAESTRO Layers -- CSA MAESTRO seven-layer taxonomy definitions for agentic AI architectures: layer identifiers (L1-L7), descriptions, keyword-to-layer mappings, and the classification algorithm (first-match-wins, "Unclassified" default). Used during Phase 1 to classify components by architectural layer.

Loading Table

Reference File Consumers Load When
references/severity-bands-shared.md orchestrator, risk-scorer, control-analyzer, threat-report, threat-infographic, report-assembler When severity classification, SLA computation, governance field derivation, or severity-based formatting is needed
references/stride-categories-shared.md orchestrator, spoofing, tampering, repudiation, info-disclosure, denial-of-service, privilege-escalation, prompt-injection, data-poisoning, model-theft, agent-autonomy, tool-abuse When category definitions, DFD applicability rules, or category-to-agent mapping is needed
references/finding-format-shared.md orchestrator, spoofing, tampering, repudiation, info-disclosure, denial-of-service, privilege-escalation, prompt-injection, data-poisoning, model-theft, agent-autonomy, tool-abuse, risk-scorer When producing or consuming finding records, validating finding structure, or assembling output tables
references/maestro-layers-shared.md orchestrator, risk-scorer, control-analyzer, threat-report When classifying components by MAESTRO layer (Phase 1), propagating layer tags through findings, or including layer metadata in output

Read the full file on GitHub · 40 lines

Files

What ships with it

7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 3d ago First seen · 40 lines · 56 tokens per session scan A 8d529d20137a

Subscribe to this mod's changes

tachi-shared is a skill published in the GitHub repository davidmatousek/tachi (90 stars, last pushed 24d ago), licensed Apache-2.0. It adds 56 tokens to every session and 724 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 2026-09-03.