Agent skill

Think About It

by Abilityai in Abilityai/cornelius

Free-form thinking mode - consider a topic through successive distinct knowledge-base perspectives (hub lenses, original frameworks, wildcard distant notes).

MITAuto-check: notesKnowledge Management

Install Think About It

skills CLI
$ npx skills add Abilityai/cornelius --skill think-about-it -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Abilityai/cornelius think-about-it --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Abilityai/cornelius.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/think-about-it .claude/skills/think-about-it && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
think-about-it
GitHub stars
109
Token cost
~2.3k tokens
SKILL.md length
897 words
Files
1
Skills in repo
54
Repo updated
First seen
Licence
MIT

At a glance

Free-form thinking mode - consider a topic through successive distinct knowledge-base perspectives (hub lenses, original frameworks, wildcard distant notes).

  • Works in 7 steps: Resolve Topic and State → Pick the Next Lens → Gather KB Material Through That Lens → …
  • Tasks that involve Knowledge bases
  • SKILL.md covers Purpose, Loop Compatibility Contract, State Dependencies and Process, plus 4 more sections
  • Calls python3

What it does

Think About It is an agent skill from Abilityai/cornelius. Free-form thinking mode - consider a topic through successive distinct knowledge-base perspectives (hub lenses, original frameworks, wildcard distant notes). One perspective pass per invocation, state persists across runs, emits [[DONE]] when the exploration weaves together. Loop-compatible by design.

Its SKILL.md is about 2.3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Knowledge Management, covering Knowledge bases. The repository describes itself as: AI-powered second brain template for Claude Code + Obsidian. The licence is MIT.

When your agent uses it

  • Tasks that involve Knowledge bases

Example prompts

  • “/think-about-it”

Requirements

  • Python 3
  • Pre-approved tools (allowed-tools): Read, Write, Edit, Grep, Glob, Bash

Workflow steps

7 steps, taken from the step headings in SKILL.md.

  1. Resolve Topic and State
  2. Pick the Next Lens
  3. Gather KB Material Through That Lens
  4. Think
  5. Append the Pass
  6. Weave Check
  7. Reply (Fixed Contract)

What it can do on your machine

Read from SKILL.md and the folder at commit b9bea90. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves these tools, so the agent can use them without asking each time:

    • Read
    • Write
    • Edit
    • Grep
    • Glob
    • Bash

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Think About It loads about 2.3k tokens when it runs. Until then it costs about 79 tokens; SKILL.md has 897 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~79
When it runs · the whole SKILL.md, loaded when a task matches
~2.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
    allowed-tools: Read, Write, Edit, Grep, Glob, Bash

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Abilityai/cornelius at commit b9bea90, republished under its MIT licence (© Abilityai). 897 words, ~2,309 tokens.

Download SKILL.mdSave it as .claude/skills/think-about-it/SKILL.md (or your agent's skills folder).
name
think-about-it
description
Free-form thinking mode - consider a topic through successive distinct knowledge-base perspectives (hub lenses, original frameworks, wildcard distant notes). One perspective pass per invocation, state persists across runs, emits [[DONE]] when the exploration weaves together. Loop-compatible by design.
allowed-tools
Read, Write, Edit, Grep, Glob, Bash
automation
autonomous
user-invocable
true
argument-hint
<topic> (free text; first run creates the thinking file, later runs continue it)

Think About It

Free-form counterpart to the incubation loop. No hypotheses, no confidence tracking, no fixed move rotation - just the same topic viewed through a different KB lens each pass, until the perspectives weave into something worth keeping.

Purpose

Put Cornelius in open thinking mode on a topic. Where /incubation-loop converges on an answer to a question, /think-about-it circles a topic - each pass looks through one distinct perspective drawn from the knowledge base and writes down what that lens reveals, distorts, or contradicts. The exploration ends with a Weave: a synthesis of the threads and candidate permanent-note seeds.

Loop Compatibility Contract

This skill is designed to be driven by a loop (/loop, /trinity:loop, or manual re-invocation):

  • One perspective pass per invocation (single-task rule; each pass completes in well under 10 minutes)
  • All state lives in the topic file - safe across separate context windows; no in-session memory assumed
  • Fixed reply format (Step 7) so {{previous_response}} chaining and Until-mode sentinels work
  • [[DONE]] appears in the reply if and only if the topic file's status is woven - tied to verifiable file state, not self-assessment
  • Idempotent on woven topics - re-invoking a finished topic re-reports the Weave and [[DONE]] without writing anything

State Dependencies

SourceLocationReadWriteDescription
Topic fileBrain/05-Meta/Thinking/Freeform/[slug].md✓✓Perspective passes + frontmatter state
Local Brain Searchresources/local-brain-search/✓Semantic search (--no-track)
Permanent notesBrain/02-Permanent/✓Evidence + wildcard lens source
MOCsBrain/03-MOCs/✓Hub lens grounding

Deliberately NOT in THINKING-REGISTRY.md - freeform topics are invisible to the incubation loop. The two engines never touch each other's state. If a Weave concludes the topic deserves rigorous treatment, promote it manually via /manage-thinking-topics.

Process

Step 1: Resolve Topic and State

$ARGUMENTS = the topic, free text. Slugify to kebab-case.

bash
date '+%Y-%m-%d'
ls "Brain/05-Meta/Thinking/Freeform/" 2>/dev/null
  • File Brain/05-Meta/Thinking/Freeform/[slug].md missing → create it from the template below (pass_count: 0, status: exploring), then continue.
  • No arguments given → list existing freeform topics with their status and pass counts, exit cleanly (no [[DONE]]).
  • File exists with status: woven → reply with the Weave summary and end with [[DONE]]. Write nothing.
Step 2: Pick the Next Lens

Read lenses_used from frontmatter. Choose exactly ONE unused lens from the deck:

Hub lenses (the KB's thematic centers of gravity): Dopamine & Motivation · Flow & Peak Performance · Buddhism & Consciousness · AI Agents & Architecture · Decision-Making & Biases · Identity & Belief Systems · Attention Economics

Framework lenses (the user's original frameworks - view the topic from inside one): Uncertainty-Dopamine-Belief Loop · The Folder Paradigm · Cross-Domain Consilience · External Anchor Principle · Self-Construction Stack · Three Gaps Framework · Four Pillars of Deep Agency · Safety as Velocity

Wildcard lens (auto-discovery style - a random permanent note from an unrelated cluster):

bash
python3 -c "import random,glob; print(random.choice(glob.glob('Brain/02-Permanent/*.md')))"

Re-roll once if the sampled note is obviously about the topic itself; otherwise the friction is the point.

Selection rules:

  • Never reuse a lens within a topic.
  • Prefer the unused lens with the most generative friction - not the most obviously related one. Obvious relevance produces summary; friction produces thought.
  • By pass 4 at least one wildcard must have been used; if not, this pass is a wildcard.
Step 3: Gather KB Material Through That Lens
bash
# --no-track: autonomous reads must not train q-values (learning hygiene, same as incubation-loop)
# read role: reasoning (contract: scope-mount) - the reasoning mount, so a lens can land on what the user has READ, not only what he wrote
BRAIN_READ_SCOPE=core,Books,document-insights python3 resources/local-brain-search/search.py "[topic] [lens core terms]" --limit 6 --mode spreading --no-track 2>/dev/null
BRAIN_READ_SCOPE=core,Books,document-insights python3 resources/local-brain-search/search.py "[lens core concept]" --limit 4 --mode spreading --no-track 2>/dev/null

Read the 3-4 most promising notes in full. These ground the pass - every pass must cite real notes, and a Books/ or Document Insights/ note is cited as encountered material, never as the user's view.

Show full SKILL.md (370 more words)Show less
Step 4: Think

Write 2-4 paragraphs of actual free-form thinking: What does the topic look like from inside this lens? What does this lens make visible that previous passes missed? What does it distort or refuse to see? Capture analogies, tensions, reframings, open questions.

  • No hypotheses, no confidence percentages - that is the incubation loop's job.
  • If a thread contradicts an earlier pass, name the contradiction explicitly and leave it standing. Premature resolution kills the weave.
  • Preserve the user's framing vocabulary where the notes supply it.
Step 5: Append the Pass

Append to the topic file:

markdown
### Pass [N]: Through the lens of [Lens Name] - YYYY-MM-DD

**Notes consulted:** [[Note A]], [[Note B]], [[Note C]]

**What this lens reveals:**
[2-4 paragraphs from Step 4]

**Threads pulled:**
- [tension / analogy / question worth keeping - 1 line each]

**Novelty:** high | medium | low

Rate novelty honestly: high = a genuinely new thread appeared; medium = deepened an existing thread; low = mostly re-derived prior passes. The honest low is what lets the loop terminate.

Update frontmatter: pass_count, lenses_used, last_two_novelty, updated, updated_by.

Step 6: Weave Check

Weave when EITHER:

  • pass_count >= 7, OR
  • the last two passes are both novelty: low (the lenses have stopped paying)

If weaving, append:

markdown
---

## The Weave

[3-6 paragraphs braiding the threads: the strongest cross-lens pattern, the contradiction
that stayed productive, what the topic turned out to actually be about]

**Candidate permanent-note seeds:**
- [2-3 one-line seeds that could become atomic notes]

**Promote to incubation?** [yes + suggested central question | no + why not]

Set status: woven in frontmatter.

Step 7: Reply (Fixed Contract)
  • Not woven: Pass N - lens: [lens]. [The single most interesting thread, 1-2 sentences.] Continuing.
  • Woven: 2-3 sentence Weave summary + the candidate note seeds + final line [[DONE]]

The reply ends with [[DONE]] if and only if status is woven.

Looping Recipes

Remote (Trinity), Until mode - the demo shape:

/trinity:loop git pull --rebase, then run /think-about-it "<topic>", then commit and push;
relay the skill's reply verbatim — until done, max 10, every 30s

(stop_signal: [[DONE]], max_runs: 10, delay_seconds: 30, timeout_per_run: 1200)

Local dynamic loop: /loop run /think-about-it "<topic>"; stop when the reply contains [[DONE]]

Manual: just invoke /think-about-it "<topic>" repeatedly; each call advances one pass.

Topic File Template

markdown
---
created: YYYY-MM-DD
updated: YYYY-MM-DD
created_by: [model-name]
updated_by: [model-name]
agent_version: 01.25
topic: "[topic as given]"
pass_count: 0
lenses_used: []
last_two_novelty: []
status: exploring
---

# Thinking Freely: [Topic Title]

## Topic
[The topic as given, one sentence of framing if needed]

---
*[Perspective passes appended below by /think-about-it]*

Error Handling

ErrorRecovery
Freeform/ directory missingCreate it, continue
KB search returns emptyRetry with alternate terms from the lens; if still empty, ground the pass in the wildcard note alone and say so in the pass
All lenses in the deck used before weaveForce weave this pass regardless of novelty
Topic file corrupt / unparseable frontmatterRename to [slug]-corrupt-YYYY-MM-DD.md, recreate fresh, note the reset in the new file

Completion Checklist

  • Topic file resolved (created or loaded)
  • One unused lens selected per the rules
  • KB searched through the lens; 3-4 notes read in full
  • Pass appended with notes cited and honest novelty rating
  • Frontmatter state updated
  • Weave check performed; Weave written if triggered
  • Reply follows the fixed contract ([[DONE]] iff woven)

© Abilityai, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/think-about-it of Abilityai/cornelius.

Open the folder on GitHubat commit b9bea90

Compare with similar skills

Think About It next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Think About It compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Think About It this skillAbilityai/cornelius109—~2.3kAutomated safety check: NotesMIT
Capture Conversationoutline/outline41k—~474Automated safety check: PassCustom licence
Find And Citeoutline/outline41k—~537Automated safety check: PassCustom licence
Xhs Virtual Productchenjin-cmd/xhs-virtual-product726—~862Automated safety check: PassMIT
OpenkbVectifyAI/OpenKB4.8k1 repos~2kAutomated safety check: WarnApache-2.0
Learn From Materialsdmoshehun-prog/learn-from-materials937—~7.9kAutomated safety check: PassMIT

Similar skills

  • Capture Conversation

    outline/outline

    Save the current conversation, a decision, or a set of notes as a document in an Outline collection; use when the user wants to keep what was discussed in their knowledge base.

    41k GitHub stars~474 tokensUpdated today
    Knowledge ManagementAuto-check passed
  • Find And Cite

    outline/outline

    Answer questions from the Outline knowledge base with quotes and links to the source documents; use when the user asks what the team knows, documented, or decided about a topic.

    41k GitHub stars~537 tokensUpdated today
    Knowledge ManagementAuto-check passed
  • Xhs Virtual Product

    chenjin-cmd/xhs-virtual-product

    This skill helps plan, select, produce, and market Xiaohongshu (RED) virtual/digital products — templates, knowledge bases, test tools, study materials.

    726 GitHub stars~862 tokensUpdated 8 days ago
    Knowledge ManagementAuto-check passed
  • Openkb

    VectifyAI/OpenKB

    A skill your agent uses when the user asks about content in their OpenKB knowledge base — research topics, concepts compiled from their documents, cross-document synthesis — or mentions openkb, an…

    4.8k GitHub starsUsed in 1 repo~2k tokens
    Knowledge ManagementAuto-check: warnings
  • Learn From Materials

    dmoshehun-prog/learn-from-materials

    Turns books, PDFs, slides and web pages into a source-grounded knowledge base and an interactive learning page in English or Chinese, with quizzes, relationship maps and reusable methodology notes.

    937 GitHub stars~7.9k tokensUpdated 2 days ago
    Knowledge ManagementAuto-check passed
  • Onyx CLI

    onyx-dot-app/onyx

    Query the Onyx knowledge base using the onyx-cli command. An agent skill from onyx-dot-app/onyx.

    32k GitHub starsUsed in 1 repo~2.6k tokens
    Knowledge ManagementAuto-check passed

More from Abilityai/cornelius

All 54 skills in this repo
  • Nano Banana Image Generator

    Abilityai/cornelius

    Generate images using Google's Nano Banana (Gemini 2.5 Flash Image).

    109 GitHub stars~1.2k tokensUpdated today
    Auto-check: notes
  • Benchmark Memory

    Abilityai/cornelius

    Systematic benchmarking framework for Local Brain Search memory system with LLM-as-judge scoring

    109 GitHub stars~3.1k tokensUpdated today
    Auto-check: notes
  • Changelog Protocol

    Abilityai/cornelius

    Protocol for creating dated changelog files after significant agent sessions.

    109 GitHub stars~555 tokensUpdated today
    Auto-check passed
  • Create Article

    Abilityai/cornelius

    Create long-form articles from knowledge base insights. An agent skill from Abilityai/cornelius.

    109 GitHub stars~3k tokensUpdated today
    Auto-check: notes
  • Decide

    Abilityai/cornelius

    Structure a decision, not just advise on it. An agent skill from Abilityai/cornelius.

    109 GitHub stars~2.7k tokensUpdated today
    Auto-check: notes
  • Epistemic Classification

    Abilityai/cornelius

    Framework for distinguishing research findings from hypotheses and speculative synthesis.

    109 GitHub stars~1.5k tokensUpdated today
    Auto-check passed

Questions about Think About It

What does Think About It do?

Free-form thinking mode - consider a topic through successive distinct knowledge-base perspectives (hub lenses, original frameworks, wildcard distant notes). Think About It is an agent skill from Abilityai/cornelius. Free-form thinking mode - consider a topic through successive distinct knowledge-base perspectives (hub lenses, original frameworks, wildcard distant notes).

When should I use Think About It?

Think About It fits situations like: tasks that involve Knowledge bases.

How do I install Think About It in Claude Code?

Run `npx skills add Abilityai/cornelius --skill think-about-it -a claude-code`. Or copy the skill folder (.claude/skills/think-about-it in Abilityai/cornelius) into .claude/skills/think-about-it in your project. Claude Code loads it when a task matches its description.

How do I install Think About It in Codex?

Run `npx skills add Abilityai/cornelius --skill think-about-it -a codex`. Or copy the skill folder (.claude/skills/think-about-it in Abilityai/cornelius) into .agents/skills/think-about-it in your project. Codex loads it when a task matches its description.

Can I use Think About It in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Abilityai/cornelius --skill think-about-it -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/think-about-it, .gemini/skills/think-about-it, .github/skills/think-about-it and .opencode/skills/think-about-it in your project.

What does Think About It need to run?

Going by SKILL.md and its folder, Think About It needs the command-line tools its instructions call (python3). Our summary lists: Python 3. Its frontmatter pre-approves these tools: Read, Write, Edit, Grep, Glob, Bash.

Does Think About It access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Think About It safe to install?

Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Think About It use?

Think About It is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Think About It use?

About 2.3k tokens (SKILL.md is roughly 9.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Think About It?

Skills that share tags, products or a category with Think About It: Capture Conversation (outline/outline, 41k stars), Find And Cite (outline/outline, 41k stars), Xhs Virtual Product (chenjin-cmd/xhs-virtual-product, 726 stars) and Openkb (VectifyAI/OpenKB, 4.8k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Think About It?

Abilityai (a GitHub organization) maintains it in Abilityai/cornelius, which has 109 GitHub stars. The repository holds 54 skills in this directory. The repository was last updated on October 8, 2026.

Source: Abilityai/cornelius on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.