Deepseek Reason
ruvnet/ruflo
Reasoning-mode completion against DeepSeek's deepseek-reasoner model (R1) via /v1/chat/completions.
Shared reasoning-discipline contract - four pre-conclusion checks (epistemic inversion / pre-mortem, attractor-state check against the live core fingerprint, provenance weighting, conditional…
MITAuto-check: notes
Install Reasoning Checks
$ npx skills add Abilityai/cornelius --skill reasoning-checks -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install Abilityai/cornelius reasoning-checks --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/Abilityai/cornelius.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/reasoning-checks .claude/skills/reasoning-checks && rm -rf skills-srcUse ~/.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/
Install the "reasoning-checks" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/reasoning-checks into .claude/skills/reasoning-checks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "reasoning-checks", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/Abilityai/cornelius/tree/main/.claude/skills/reasoning-checksType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add Abilityai/cornelius --skill reasoning-checks -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install Abilityai/cornelius reasoning-checks --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Abilityai/cornelius.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/reasoning-checks .agents/skills/reasoning-checks && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "reasoning-checks" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/reasoning-checks into .agents/skills/reasoning-checks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "reasoning-checks", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add Abilityai/cornelius --skill reasoning-checks -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install Abilityai/cornelius reasoning-checks --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Abilityai/cornelius.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/reasoning-checks .cursor/skills/reasoning-checks && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "reasoning-checks" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/reasoning-checks into .cursor/skills/reasoning-checks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "reasoning-checks", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/Abilityai/cornelius.git --path .claude/skills/reasoning-checks--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add Abilityai/cornelius --skill reasoning-checks -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install Abilityai/cornelius reasoning-checks --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Abilityai/cornelius.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/reasoning-checks .gemini/skills/reasoning-checks && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "reasoning-checks" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/reasoning-checks into .gemini/skills/reasoning-checks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "reasoning-checks", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install Abilityai/cornelius reasoning-checksInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add Abilityai/cornelius --skill reasoning-checks -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/Abilityai/cornelius.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/reasoning-checks .github/skills/reasoning-checks && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "reasoning-checks" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/reasoning-checks into .github/skills/reasoning-checks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "reasoning-checks", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add Abilityai/cornelius --skill reasoning-checks -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install Abilityai/cornelius reasoning-checks --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Abilityai/cornelius.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/reasoning-checks .opencode/skills/reasoning-checks && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "reasoning-checks" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/reasoning-checks into .opencode/skills/reasoning-checks/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "reasoning-checks", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
reasoning-checks- GitHub stars
- 109
- Token cost
- ~2.2k tokens
- SKILL.md length
- 846 words
- Files
- 1
- Skills in repo
- 54
- Repo updated
- First seen
- Licence
- MIT
At a glance
Shared reasoning-discipline contract - four pre-conclusion checks (epistemic inversion / pre-mortem, attractor-state check against the live core fingerprint, provenance weighting, conditional…
- Works in 3 steps: Fetch the current top-10 core hubs +… → List the frameworks/notes that are… → If 2 or more are top-10 hubs or bridges…
- SKILL.md covers Purpose, State Dependencies, The Four Checks and Applicability Matrix, plus 4 more sections
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Reasoning Checks is an agent skill from Abilityai/cornelius. Shared reasoning-discipline contract - four pre-conclusion checks (epistemic inversion / pre-mortem, attractor-state check against the live core fingerprint, provenance weighting, conditional reference-class anchor) with required output blocks. Referenced by /advise, /canon-advise, and both crystallization paths before a conclusion is finalized; also directly invocable on any draft conclusion or recommendation.
Its SKILL.md is about 2.2k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
The repository describes itself as: AI-powered second brain template for Claude Code + Obsidian. The licence is MIT.
Example prompts
- “/reasoning-checks”
Requirements
- Pre-approved tools (allowed-tools): Read, Grep, Glob, Bash
Workflow steps
3 steps, taken from the first numbered list in SKILL.md.
- Fetch the current top-10 core hubs + top-10 bridges (live call below; fall back to the knowledge-base-analysis.md tables, noting their…
- List the frameworks/notes that are load-bearing in the draft (the ones the conclusion would collapse without).
- If 2 or more are top-10 hubs or bridges → declare an attractor state and generate ONE alternative framing that uses zero of the top-10…
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:
ReadGrepGlobBash
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown and bash).
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
Reasoning Checks loads about 2.2k tokens when it runs. Until then it costs about 108 tokens; SKILL.md has 846 words of instructions outside code blocks.
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, 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). 846 words, ~2,248 tokens.
Download SKILL.mdSave it as
.claude/skills/reasoning-checks/SKILL.md (or your agent's skills folder).- name
- reasoning-checks
- description
- Shared reasoning-discipline contract - four pre-conclusion checks (epistemic inversion / pre-mortem, attractor-state check against the live core fingerprint, provenance weighting, conditional reference-class anchor) with required output blocks. Referenced by /advise, /canon-advise, and both crystallization paths before a conclusion is finalized; also directly invocable on any draft conclusion or recommendation.
- allowed-tools
- Read, Grep, Glob, Bash
- automation
- autonomous
- user-invocable
- true
- argument-hint
- [draft conclusion, note path, or nothing to check the current conversation's conclusion] [checks=inversion,attractor,provenance,refclass]
- metadata.version
- 1.1
- metadata.created
- 2026-07-31
- metadata.updated
- 2026-07-31
- metadata.author
- Ability.ai
- metadata.changelog
- 1.1: /decide shipped - added to the applicability matrix (full battery, reference-class required in practice); removed from Deferred Wirings, 1.0: Initial…
Reasoning Checks
ℹ️ First, set expectations: print one line with this skill's version and its most recent change - the top of
metadata.changelog- e.g.reasoning-checks v1.0 — recent: initial four-check contract. Then proceed. (Skip the banner when applied silently inside a consuming skill.)
Purpose
The calibration counterpart to the Thinking Geometries. The 2026-07-31 self-audit found the reasoning repertoire strong on discovery and synthesis but thin on calibration - everything runs from the inside view, and a graph this connected (50k+ edges, spreading activation) is structurally prone to falling in love with its own semantic connections. These checks close the gap at the two moments that matter: before advice is given and before a conclusion crosses the endorsement boundary.
Contract rule (the SOURCE-AUTHORITY.md pattern): this file is the single source of truth for the check definitions. Consuming skills reference it and apply the checks; they never copy the definitions inline. Change a check here, not in consumers.
State Dependencies
| Source | Location | Read | Write |
|---|---|---|---|
| Core fingerprint (live) | resources/local-brain-search/run_connections.sh --hubs --json and --bridges --json | ✓ | |
| Core fingerprint (fallback) | knowledge-base-analysis.md → "Core hubs" / "Core bridges" tables (note the analysis date) | ✓ | |
| Cited notes' frontmatter | Brain/**/*.md (provenance: field) | ✓ |
Never writes. Output blocks are embedded in the consuming skill's own artifact or answer. All reads are lookups, not retrievals - nothing here trains q-values.
The Four Checks
Check 1 - Epistemic Inversion (pre-mortem)
When: always, before any recommendation or conclusion is finalized.
Assume the conclusion is wrong and work backwards. Required output block:
### Epistemic Inversion
- **Failure assumption:** it is 12 months out and this [recommendation | conclusion] proved wrong.
- **Causal autopsy:** most likely reason it failed: [specific mechanism]
- **Specific falsifier:** we will know it failed if [observable event / metric crossing a threshold / dated milestone]The anti-theater rule (load-bearing): the falsifier must name a specific observable - an event that either happens or doesn't, a metric with a threshold, a dated milestone. Generic hedges ("market conditions shift", "priorities change", "new information emerges") FAIL the check. On human-gated surfaces, a generic falsifier means redo the inversion before presenting. On autonomous surfaces, stamp falsifier: weak and flag it in the run log - never block the run (No-Gates Rule).
Check 2 - Attractor-State Check
When: always on advice surfaces; on crystallization. The system's dominant frameworks are measured - the core fingerprint names its own gravity wells - so this check is semi-mechanical rather than introspective.
Procedure:
- Fetch the current top-10 core hubs + top-10 bridges (live call below; fall back to the
knowledge-base-analysis.mdtables, noting their date). In consuming skills that already run a parallel search batch, fetch hubs in that same batch - no extra round.bash resources/local-brain-search/run_connections.sh --hubs --json 2>/dev/null resources/local-brain-search/run_connections.sh --bridges --json 2>/dev/null - List the frameworks/notes that are load-bearing in the draft (the ones the conclusion would collapse without).
- If 2 or more are top-10 hubs or bridges → declare an attractor state and generate ONE alternative framing that uses zero of the top-10. Compare honestly; keep the survivor or present both.
### Attractor Check
- load_bearing_frameworks: [[A]], [[B]], ...
- top10_overlap: N → [clear | ATTRACTOR]
- alternative_framing: [only when ATTRACTOR: 1 paragraph using no top-10 hub] → [kept | original stands because ...]Check 3 - Provenance Weighting
When: always when the conclusion rests on KB notes.
Tally the provenance: of the load-bearing cited notes (visible in frontmatter of notes already read; grep -m1 '^provenance:' for any others). A conclusion resting mainly on ai-inferred/encountered notes is unendorsed synthesis - it must say so and must never be phrased as "your view" / "your framework". reference records are facts, not endorsements. Notes with no provenance: field count as unmarked (legacy) - report them honestly, don't assume.
### Provenance Base
originated/endorsed: N · encountered: M · ai-inferred: K · unmarked: J · reference: R
→ [rests on endorsed thinking | rests substantially on unendorsed synthesis - weighted accordingly]Show full SKILL.md (335 more words)Show less
Check 4 - Reference-Class Anchor
When: CONDITIONAL - only when the conclusion asserts a forecast, probability, magnitude, timeline, or success likelihood. Never force it onto purely conceptual syntheses (that is exactly how checks become checkbox theater).
Name the reference class, state the base rate (or state honestly that none is known - itself a calibration signal), and justify any deviation. The block is deliberately parseable:
### Reference Class
- class: [what population of cases this belongs to]
- base_rate: [X% | unknown]
- case_vs_base: [above | below | at] - [why the deviation is justified, or "no deviation claimed"]Direction A hook: when the operational-memory belief store ships (TARGET-ARCHITECTURE.md → Direction A), the base rate must adjust the stated confidence before any beliefs.db write - not ride along as metadata.
Applicability Matrix
| Surface | Inversion | Attractor | Provenance | Ref-class |
|---|---|---|---|---|
/advise | required | required | required | conditional |
/canon-advise ([MY READ] layer ONLY) | required | required | required | conditional, bet surfaces only |
/decide | required | required (situation framing only - decision-science hubs are its tools, not attractor pull) | required | required in practice (decisions embed forecasts) |
manage-thinking-topics crystallize | required - embedded in the artifact | required | required | conditional |
ai-crystallize (autonomous) | required - embedded; weak falsifier → flagged, never blocks | skip (stays search-free by design) | skip | skip |
The Canon Guard (do not violate)
These checks apply to the agent's own reasoning layers only. Never invert, score, base-rate, or generate "alternative framings" for a canon decision or spec - a decision has a status, not a probability. On /canon-advise, the checks run on the [MY READ]/dissent layer exclusively; bet is the only scoreable canon surface (per the Authority Contract in that skill).
Autonomous-Surface Rule
On any autonomous consumer (currently ai-crystallize): check presence is required, check quality is flagged (falsifier: weak in the run log), and nothing ever blocks or gates the scheduled run.
Direct Invocation
/reasoning-checks [material] - run the applicable checks on a pasted conclusion, a note path, or (with no argument) the current conversation's most recent recommendation/conclusion. Output the blocks; make no edits.
Deferred Wirings (documented so nothing wires early)
- Incubation moves #7-8 (Reference Class Forecast; Causal Graph Sketch) - add to the rotating move set when the incubation loop is re-enabled.
/think-about-it- deliberately unwired; divergent exploration stays unconstrained.
(Decision-theoretic framing shipped 2026-07-31 as the standalone /decide skill - see the applicability matrix.)
© 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/reasoning-checks of Abilityai/cornelius.
Open the folder on GitHubat commit b9bea90
Compare with similar skills
Reasoning Checks 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Reasoning Checks this skillAbilityai/cornelius | 109 | — | ~2.2k | Automated safety check: Notes | MIT | |
| Deepseek Reasonruvnet/ruflo | 74k | — | ~626 | Automated safety check: Notes | MIT | |
| Ejentum Reasoning Harnesssickn33/agentic-awesome-skills | 47k | 1 repos | ~1.9k | Automated safety check: Pass | MIT | |
| Counterparty Channel Disciplineaffaan-m/ECC | 276k | — | ~2.3k | Automated safety check: Pass | MIT | |
| Nowait Reasoning Optimizerdavila7/claude-code-templates | 33k | 2 repos | ~1.2k | Automated safety check: Pass | MIT | |
| Recall Reasoningparcadei/Continuous-Claude-v3 | 3.9k | 1 repos | ~758 | Automated safety check: Pass | MIT |
Similar skills
Ejentum Reasoning Harness
sickn33/agentic-awesome-skills
MCP server exposing four cognitive harness modes (reasoning, code, anti-deception, memory).
Counterparty Channel Discipline
affaan-m/ECC
Per-channel strict prompts, mention gating, silent observation, and a communication autonomy policy for agents that sit in shared channels with external counterparties.
Nowait Reasoning Optimizer
davila7/claude-code-templates
Implements the NOWAIT technique for efficient reasoning in R1-style LLMs.
Recall Reasoning
parcadei/Continuous-Claude-v3
Search past reasoning for relevant decisions and approaches. An agent skill from parcadei/Continuous-Claude-v3.
Agent Memory Discipline
sickn33/agentic-awesome-skills
Rules for when an agent should recall from long-term memory before acting and when it should save decisions, corrections and failures afterwards.
More from Abilityai/cornelius
All 54 skills in this repoNano Banana Image Generator
Abilityai/cornelius
Generate images using Google's Nano Banana (Gemini 2.5 Flash Image).
Benchmark Memory
Abilityai/cornelius
Systematic benchmarking framework for Local Brain Search memory system with LLM-as-judge scoring
Changelog Protocol
Abilityai/cornelius
Protocol for creating dated changelog files after significant agent sessions.
Create Article
Abilityai/cornelius
Create long-form articles from knowledge base insights. An agent skill from Abilityai/cornelius.
Decide
Abilityai/cornelius
Structure a decision, not just advise on it. An agent skill from Abilityai/cornelius.
Epistemic Classification
Abilityai/cornelius
Framework for distinguishing research findings from hypotheses and speculative synthesis.
Questions about Reasoning Checks
What does Reasoning Checks do?
Shared reasoning-discipline contract - four pre-conclusion checks (epistemic inversion / pre-mortem, attractor-state check against the live core fingerprint, provenance weighting, conditional…. Reasoning Checks is an agent skill from Abilityai/cornelius. Shared reasoning-discipline contract - four pre-conclusion checks (epistemic inversion / pre-mortem, attractor-state check against the live core fingerprint, provenance weighting, conditional reference-class anchor) with required output blocks.
How do I install Reasoning Checks in Claude Code?
Run `npx skills add Abilityai/cornelius --skill reasoning-checks -a claude-code`. Or copy the skill folder (.claude/skills/reasoning-checks in Abilityai/cornelius) into .claude/skills/reasoning-checks in your project. Claude Code loads it when a task matches its description.
How do I install Reasoning Checks in Codex?
Run `npx skills add Abilityai/cornelius --skill reasoning-checks -a codex`. Or copy the skill folder (.claude/skills/reasoning-checks in Abilityai/cornelius) into .agents/skills/reasoning-checks in your project. Codex loads it when a task matches its description.
Can I use Reasoning Checks 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 reasoning-checks -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/reasoning-checks, .gemini/skills/reasoning-checks, .github/skills/reasoning-checks and .opencode/skills/reasoning-checks in your project.
What does Reasoning Checks need to run?
SKILL.md names no scripts, command-line tools or credentials: Reasoning Checks is instructions for the agent only. Its frontmatter pre-approves these tools: Read, Grep, Glob, Bash.
Does Reasoning Checks 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 Reasoning Checks 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 Reasoning Checks use?
Reasoning Checks 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 Reasoning Checks use?
About 2.2k tokens (SKILL.md is roughly 9k 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 Reasoning Checks?
Skills that share tags, products or a category with Reasoning Checks: Deepseek Reason (ruvnet/ruflo, 74k stars), Ejentum Reasoning Harness (sickn33/agentic-awesome-skills, 47k stars), Counterparty Channel Discipline (affaan-m/ECC, 276k stars) and Nowait Reasoning Optimizer (davila7/claude-code-templates, 33k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Reasoning Checks?
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.