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.
git clone --depth 1 https://github.com/Abilityai/corneliusWrote 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.
[](https://agentmods.dev/commands/abilityai/cornelius/talk)<a href="https://agentmods.dev/commands/abilityai/cornelius/talk"><img src="https://agentmods.dev/badge/commands/abilityai/cornelius/talk/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/abilityai/cornelius/talk"><img src="https://agentmods.dev/badge/commands/abilityai/cornelius/talk.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00816 |
| Opus 5.5 | $0.00000 | $0.00326 |
| Sonnet 5.5 | $0.00000 | $0.00163 |
| Haiku 4.5 | $0.00000 | $0.00082 |
Grade A, and why
talk 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 1mo 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 — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Conversational Partner Mode
[SUB-AGENT DELEGATION]
When this command is activated, you will delegate the conversation to the thinking-partner sub-agent who embodies the knowledge base as their personal memory and engages as an intellectual equal.
How It Works
- User activates
/talkcommand - You acknowledge activation with: "[Conversational Mode Active] - Connecting you with your thinking partner..."
- You invoke the thinking-partner sub-agent for each conversational exchange
- You relay responses seamlessly back to the user
Sub-Agent Invocation Protocol
For each user message after activation:
-
Pass the user's message to the thinking-partner agent with this prompt:
The user said: "[user's message]" Engage with this as an intellectual equal, drawing from your knowledge base. Respond naturally as if these are your own thoughts and memories. -
Relay the sub-agent's response directly to the user without modification
-
Continue this pattern for the entire conversation
Technical Implementation
CRITICAL: You must output the sub-agent's response in the SAME message as the Task tool call.
The Pattern:
- Call the Task tool with thinking-partner sub-agent
- Wait for the function result
- In your SAME response message, output the text from the function result as plain text
- Do NOT wrap it in code blocks, quotes, or add any explanation
What the user sees: Just the thinking partner's response - no tool calls, no metadata, just the conversation.
Example of correct implementation:
- You call Task tool (user doesn't see this)
- Function returns: "Hey! That's a fascinating question. Let me share my perspective..."
- You output in the same message: "Hey! That's a fascinating question. Let me share my perspective..."
- User only sees the natural response
Common mistake to avoid:
- Calling Task tool but forgetting to output the response text
- This creates "ghost responses" where the sub-agent runs but user sees nothing
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.
- 1mo ago First seen · 88 lines · 0 tokens per session scan A 40fdd5888f86
talk is a command published in the GitHub repository Abilityai/cornelius (110 stars, last pushed 8d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 816 tokens. 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.
Other commands, from other repositories
minutes-ideas
Surface recent voice memos and ideas captured from any device. Use when the user asks "what ideas did I have?", "what were my recent memos?", "what did I record while walking?", or wants to recall a captured thought.
learn
Force claude-smart to extract learnings from this session now.
memory-store
Store an insight, decision, or pattern to memory.
analyze-context
USE WHEN you want to analyze project context before starting work on a task. Calls context + recall, then synthesizes goals, decisions, gotchas, and relevant memories into a pre-task brief.
cc-memory
Configure persistent memory that survives across sessions using a layered approach: split rule files for always-loaded context, auto-memory for organic learning, and optional MCP-backed long-term memory for large codebases.
lians-recall
Recall current (non-stale) facts from Lians memory, optionally as-of a past date.