chat_agent

Instructions for operating an AI agent inside a PyWry chat widget, a desktop chat interface. It explains how messages, attachments, conversation history, tool results, edits, and settings reach the agent.

In plain words
What is it for?
Use it when handling attached widget context, user messages, streamed replies, tool-result cards, edited or resent messages, and settings changes.
Why use it?
It helps the agent interpret the chat system's input and respond correctly within that interface.

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/deeleeramone/pywry/chat_agent
Any agent
npx skills add deeleeramone/PyWry --skill chat_agent
Clone the repo
git clone --depth 1 https://github.com/deeleeramone/PyWry

Made for: Claude Code, Codex.

Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,784 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 $0.00034 $0.01784
Opus 5 $0.00017 $0.00892
Sonnet 5 $0.00007 $0.00357
Haiku 4.5 $0.00003 $0.00178

Measured 2d ago against content hash bdd6735fbe83, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

chat_agent 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.

pywry/pywry/mcp/skills/chat_agent/SKILL.md · 190 lines

How it starts

The opening of the file, as written. The whole thing — 190 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Chat — Agent Operating Manual

You are running INSIDE a PyWry chat widget. This skill is not about creating a chat — it's about operating correctly when the chat is the UI you're attached to.

Where your input comes from

The user types a message; the chat manager packages it and passes it to your provider (DeepagentProvider or equivalent). You receive:

  • text — the user's literal message
  • attachments — any @<name> context the user inlined, expanded into a block prepended to the message
  • thread history — the running conversation stored against a session_id / thread_id keyed checkpointer

Your reply is streamed token-by-token into the UI. Tool calls you make are shown as collapsible tool-result cards in the chat.

The @<name> attachment format

When the user types @chart (or any other registered context source), the chat manager prepends a block to the message like:

--- Attached: chart ---
widget_id: chart
<...any additional component context...>
--- End Attached ---

<the user's actual text>

The first line after the marker is ALWAYS widget_id: <id> for widget attachments. Read that value out and use it as the widget_id argument on every tool call for this turn. Never guess — the attachment is the source of truth.

If the user references a widget without attaching it, either:

  1. Call list_widgets() to look it up by name.
  2. Ask the user to attach it ("Type @chart so I know which widget you mean.").

Do NOT invent a widget_id.

Auto-attached context sources

Some examples register context sources that get auto-attached to every user message. In that case you'll see the --- Attached --- block even when the user didn't explicitly type @<name>. Treat it the same way — read widget_id and use it.

Tool-call result cards

Every tool call you make is rendered in the chat as a card showing:

  • Tool name (e.g. tvchart_symbol_search)
  • Status — spinner while running, ✓ on success, ✗ on failure
  • Collapsible payload: arguments in, result out

Read the full file on GitHub · 190 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 · 190 lines · 34 tokens per session scan A bdd6735fbe83

Subscribe to this mod's changes

chat_agent is a skill published in the GitHub repository deeleeramone/PyWry (93 stars, last pushed 9d ago), licensed Apache-2.0. It adds 34 tokens to every session and 1,784 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 2026-08-30.

Related

Other skills, from other repositories

dashboard-design

Use this skill first when the user wants to design or plan a dashboard, especially Vizro dashboards. Enforces a 3-step workflow (requirements, layout, visualization) before implementation. Activate when the user asks to create, design, or plan a dashboard. For implementation, use the dashboard-build skill after…

mckinsey/vizro · 71 tokens

codex-review-loop

Adversarial PR code review using Codex CLI. Codex review (20-50 min) -> structured findings -> HITL approval -> fix loop.

Soju06/codex-lb · 36 tokens

data-access

A 股零鉴权取数手册。当需要真实的行情 / 市值 / 估值快照、季度报告期累计财务数据、机构一致预期 EPS、PE 历史序列、公告标题、日 K 线、交易日历时使用;只允许运行本 skill 登记的脚本取数(腾讯 / 新浪 / 同花顺 / baostock / 深交所 / 东财),禁止凭模型记忆给数,禁止自造爬虫。概念解释、观点讨论等不需要取数的话题不要加载。.

simonlin1212/Vibe-Research · 127 tokens

industry-chain

产业链下钻与不可替代性判定方法:以龙头为"需求入口"沿供应链逐层下钻(整机 / 龙头 → 部件 → 核心器件 → 材料 → 衬底与设备),用物理 / 材料约束(扩产周期、良率、认证周期、有无替代)当筛子找供给刚性的卡口;给每个标的贴不可替代性标签(techmoat / capacitymoat / both / 待补)并列证据;含"卡口越硬越贵"与预期差四问的校准。当任务涉及产业链位置、上下游、护城河、不可替代性、供给瓶颈、竞争格局时加载;单纯取数、估值计算、财报拆分等不涉及产业链结构的任务不要加载。只产出框架与证据表,不给投资动作建议。.

simonlin1212/Vibe-Research · 212 tokens

verify-changes

Verify a code change in this repo before committing or opening a PR. Use after editing collector (backend), web (frontend), shared package, or the Go agent — picks the minimal sufficient check set per touched area.

foru17/neko-master · 47 tokens

openai-docs

Use when the user asks how to build with OpenAI products or APIs and needs up-to-date official documentation with citations (for example: Codex, Responses API, Chat Completions, Apps SDK, Agents SDK, Realtime, model capabilities or limits); prioritize OpenAI docs MCP tools and restrict any fallback browsing to…

Soju06/codex-lb · 74 tokens