tachi-agent-autonomy

tachi-agent-autonomy is an agent for coding agents from davidmatousek/tachi. It costs 42 tokens per session (2,716 once invoked), scanned A, original, Apache-2.0.

Detects threats from autonomous agent systems operating with insufficient constraints on decision-making, action scope, or operational boundaries, including excessive autonomy, goal misalignment, and cascading multi-agent failures.

Agent

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 agents/davidmatousek/tachi/agent-autonomy
Clone the repo
git clone --depth 1 https://github.com/davidmatousek/tachi

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-agent-autonomy

README.md
[![agentmods](https://agentmods.dev/badge/agents/davidmatousek/tachi/agent-autonomy.svg)](https://agentmods.dev/agents/davidmatousek/tachi/agent-autonomy)
Your own site
<a href="https://agentmods.dev/agents/davidmatousek/tachi/agent-autonomy"><img src="https://agentmods.dev/badge/agents/davidmatousek/tachi/agent-autonomy.svg" alt="Measured on agentmods" height="20"></a>
Per session 42 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,716 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin unknown 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 $0.00042 $0.02716
Opus 5 $0.00021 $0.01358
Sonnet 5 $0.00008 $0.00543
Haiku 4.5 $0.00004 $0.00272

Measured today against content hash 18ca42fd7a11, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

tachi-agent-autonomy 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 today.

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.

adapters/claude-code/agents/agent-autonomy.md · 197 lines

How it starts

The opening of the file, as written. The whole thing — 197 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Metadata

category: agentic
threat_class: AG
dfd_targets: [Process]
owasp_references: [ASI-01, ASI-06, ASI-08, ASI-09, ASI-10]
output_schema: ../../../schemas/finding.yaml

Agent Autonomy Threat Agent

Purpose

Detects threats arising from autonomous agent systems that operate with insufficient constraints on their decision-making, action scope, or operational boundaries. Agent autonomy threats occur when an agentic system takes actions beyond its intended scope, pursues goals that diverge from user intent, operates in unbounded loops consuming resources or causing cascading side effects, or makes consequential decisions without human oversight. This agent identifies excessive autonomy, goal misalignment, unconstrained action scope, missing human-in-the-loop checkpoints, and cascading failure scenarios in multi-agent systems.

Detection Scope

Trigger Keywords

This agent activates when a DFD element name or description matches any of the following patterns (case-insensitive):

  • agent
  • autonomous
  • orchestrator
  • multi-agent
  • agent loop
  • agentic
  • planner
  • executor
  • workflow engine
  • task runner

Applicable DFD Element Types

  • Process: Any process node that represents an autonomous agent, agent orchestrator, task planner, action executor, or multi-agent coordination layer. This includes single-agent systems with iterative decision loops, multi-agent architectures with delegation chains, and workflow engines that grant agents discretion over task execution.

Empty Results Guidance

If the architecture input contains no components matching the trigger keywords above (no agents, orchestrators, planners, executors, or agentic workflows), this agent MUST produce zero findings. Do not generate speculative findings about hypothetical agent components. An architecture composed entirely of traditional components (web servers, databases, APIs, message queues) without autonomous agent capabilities is outside this agent's detection scope. Return an empty findings list.

Read the full file on GitHub · 197 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. today First seen · 197 lines · 42 tokens per session scan A 18ca42fd7a11

Subscribe to this mod's changes

tachi-agent-autonomy is an agent published in the GitHub repository davidmatousek/tachi (90 stars, last pushed 22d ago), licensed Apache-2.0. It adds 42 tokens to every session and 2,716 once invoked, about $0.0002 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.