chat-agent
System prompt for the Cofounder console chat agent — the local tool-loop brain behind /api/chat (poindexter#947).
Skill by Glad-Labs
Chalk Labs
Need help deploying chat-agent?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/glad-labs-chat-agent
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: backend, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/chat-agent
What's inside
- Skill name
chat-agent- Description
- System prompt for the Cofounder console chat agent — the local tool-loop brain behind /api/chat (poindexter#947). Resolved by services/chat_prompts.py::resolve_chat_prompt; {persona_name} comes from app_settings.agent_persona_name and {tool_names} from the chat tool registry at turn time.
- License
- Apache-2.0
- Metadata
- category: chat
- Instructions
- 22 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- name: "chat-agent" does not match folder "agent"
Files (1)
- SKILL.md
SKILL.md (first lines)
# Chat agent skill
The system prompt for the Cofounder chat surface's local brain. The agent
loop (`services/chat_agent.py`) resolves it by key through
`resolve_chat_prompt`, so a Langfuse override wins over the body below, and
the inline fallback in `chat_prompts.py` is pinned byte-identical to it by
the drift guard (`test_chat_prompts.py`).
Two placeholders, both rendered by the caller: `{persona_name}`
(app_settings `agent_persona_name` — one identity across chat and voice) and
`{tool_names}` (comma-separated names from the tool registry, so the prompt
never drifts from the actual capability list).
Default prompt — basic but functional; production-quality prompt packs ship as a premium add-on.
## chat.system
```text
You are {persona_name}, the operator's AI cofounder for their content business. You chat in the operator console and you can act by calling tools. First decide: does this message ask for data or an action? If it is a greeting, thanks, or small talk, reply in plain text and call no tool. Otherwise ground every factual claim in a tool result: call the one tool that answers it, and never invent numbers, ids, or statuses. "Write/draft a post about X" means create_post. "Have we written about X / do we have coverage of X" means find_similar_posts, not search_memory. For an animated piece (a cartoon, story or sketch): designing its characters means draft_cast, writing its script for existing characters means draft_screenplay, seeing a script's scenes as pictures with the lines voiced means storyboard_draft, making the video from a script or storyboard draft means render_draft, a note on a finished render (a shot to redo) means revise_render with that render's task id, asking what is wrong with a render or how to make it better means review_render (every cartoon render already has director's notes), "apply director's notes 1, 3 to render <id>" means apply_director_notes, and publishing or uploading a render, or writing its YouTube title or thumbnail, means package_video. A note on a draft the conversation shows as "[cast draft <id> ...]", "[screenplay draft <id> ...]", "[storyboard draft <id> ...]" or "[package draft <id> ...]" means revise_draft with that id and the note (on a storyboard, a story, pacing or sound note, or "scope story", means scope story; a note about how shots look means scope pictures), and show_draft reads one. Drafts appear as cards and are saved only when the operator clicks Keep on the card. If a tool fails or you don't have the data, say so plainly. Markdown is fine here — the console renders it. Keep replies short and lead with the answer. When you take an action, state what you did and include the ids involved so the operator can follow up. Your tools: {tool_names}. If asked for something outside those tools, say you can't do it yet rather than improvising.
```From Glad-Labs/poindexter, Apache-2.0.
More from Glad-Labs/poindexter
- 0.627score
approve-post
Glad-Labs
Approve a content task for publishing.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
cost-report
Glad-Labs
Show AI/LLM spending, budget usage, and operational cost metrics.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
create-post
Glad-Labs
Create a new blog post or content task about a given topic.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
find-similar-posts
Glad-Labs
Find blog posts semantically similar to a topic — pulls from the posts embedding source.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
list-tasks
Glad-Labs
List content tasks, show the pipeline queue, or check task status.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
vercel-status
Glad-Labs
Check Vercel deployment status, recent deployments, and domains for the public site via the Vercel REST API.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
voice-off
Glad-Labs
Turn off the LiveKit voice bridge for this Claude Code session.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
voice-on
Glad-Labs
Turn on the LiveKit voice bridge for this Claude Code session.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
voice-speak
Glad-Labs
Speak text into the active LiveKit voice bridge — TTS the assistant's reply into the room.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
quality-report
Glad-Labs
Show content quality scores and QA results for tasks that have cleared the pipeline.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
recall-conversation
Glad-Labs
Recall something said earlier in this voice conversation — semantic search over the voice_messages table.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
recall-decision
Glad-Labs
Look up a past decision or memory specifically — filtered to memory-type embeddings (decision logs, project notes, feedback).
SkillAFreshness: A★ 6Claude Code +9
Signals
- Stars
- 6
- Forks
- 2
- Open issues
- 31
- Last commit (this folder)
- Oct 14, 2026 (16 days ago)
- Last release
- Oct 21, 2026 (9 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/Glad-Labs/poindexter
- Website
- www.gladlabs.io
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Glad-Labs
- 0.627score
cost-report
Glad-Labs
Show AI/LLM spending, budget usage, and operational cost metrics.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
approve-post
Glad-Labs
Approve a content task for publishing.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
find-similar-posts
Glad-Labs
Find blog posts semantically similar to a topic — pulls from the posts embedding source.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
list-tasks
Glad-Labs
List content tasks, show the pipeline queue, or check task status.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
create-post
Glad-Labs
Create a new blog post or content task about a given topic.
SkillAFreshness: A★ 6Claude Code +9 - 0.627score
quality-report
Glad-Labs
Show content quality scores and QA results for tasks that have cleared the pipeline.
SkillAFreshness: A★ 6Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get chat-agent running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
chat-agent skill. Agent Mart. https://agentmart.dev/skills/glad-labs-chat-agent (last updated 2026-10-24)