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.
npx agentmods add skills/gettalon/talon-plugins/create-workflownpx skills add gettalon/talon-plugins --skill create-workflowgit clone --depth 1 https://github.com/gettalon/talon-pluginsWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00052 | $0.01270 |
| Opus 5 | $0.00026 | $0.00635 |
| Sonnet 5 | $0.00010 | $0.00254 |
| Haiku 4.5 | $0.00005 | $0.00127 |
Grade A, and why
create-workflow 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.
How it starts
The opening of the file, as written. The whole thing — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create Workflow — State Machine Designer
Design a YAML workflow that enforces a multi-step process with state transitions, per-state tool restrictions, and permission modes.
When to Use Workflows vs Skills
| Use a Skill | Use a Workflow |
|---|---|
| Repeatable task with steps | Process with enforced stages |
| Agent follows instructions voluntarily | System enforces what tools are available per stage |
| No state tracking needed | State tracked across interactions |
| e.g., "Run a DCF valuation" | e.g., "Plan → Implement → Review → Ship" |
Rule of thumb: If the agent must NOT skip a stage or use certain tools before completing a prior stage, use a workflow. If it's guidance the agent follows, use a skill.
Process
Step 1: Understand the Process
Ask one at a time:
-
What process does this workflow enforce?
- e.g., "Code change: plan → implement → review → verify"
- e.g., "Research: gather → analyze → validate → present"
- e.g., "Incident: triage → investigate → remediate → postmortem"
-
What are the stages (states)?
- List each stage and what happens there
- What marks a stage as complete (transition condition)?
-
Should tools be restricted per stage?
- e.g., Planning stage: read-only tools only
- e.g., Review stage: no file writes
- e.g., Implementation: all tools available
-
What permission mode per stage?
plan— read-only, agent proposes but doesn't actask— agent asks before each actionallow— agent can read/write freelybypass— full autonomy including shell execution
Step 2: Design the State Machine
Map out states and transitions:
[initial] → [state-2] → [state-3] → [done]
↑ |
└────── (rejected) ──────┘
For each state, define:
- name: Short identifier (snake_case)
- allowed_tools: Tool allowlist (null = all)
- permission_mode: plan | ask | allow | bypass (optional, derived from tools if omitted)
- on_entry: Instructions when entering this state
- on_exit: Cleanup when leaving (optional)
- transitions: List of possible next states with optional conditions
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.
- 2d ago First seen · 177 lines · 52 tokens per session scan A 6ddec6ee0da1
create-workflow is a skill published in the GitHub repository gettalon/talon-plugins (6 stars, last pushed 5mo ago), licensed MIT. It adds 52 tokens to every session and 1,270 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 2026-08-31.
Other skills, from other repositories
browser
Use this skill when the user says browser, /browser, test in Chrome, inspect a webpage, verify a localhost app, capture screenshots, check console/network errors, run browser QA, or automate browser flows with the Mochi browser MCP.
browse
Use when asked to interact with a web browser — navigate, click, fill forms, take screenshots, run JavaScript, or read page content. Requires the browser-bridge MCP server and Chrome extension.
NanoBanana-PPT-Skills
Skill "NanoBanana-PPT-Skills" from girish6055/NanoBanana-PPT-Skills, covering ppt generator pro - claude code skill, 📋 元数据, ✨ 功能特性, 核心功能 and 新功能 (v2.0).
codex-browser
Control Chrome or Edge via the ChatGPT desktop app's browser bridge (formerly Codex Desktop). 52 MCP tools for tabs, navigation, DOM, input, CDP, network, file upload, dialog handling, and diagnostics.
browser-use
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work.
remote-browser
Controls an isolated Browser Use Cloud browser from a sandboxed machine with the current Browser Use CLI.