Install the "ref-reconcile" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/ref-reconcile into .claude/skills/ref-reconcile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ref-reconcile", 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.
Type 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.
skills CLI
$ npx skills add Abilityai/cornelius --skill ref-reconcile -a codex
Project install goes to .agents/skills/; add -g for ~/.codex/skills/.
Install the "ref-reconcile" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/ref-reconcile into .agents/skills/ref-reconcile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ref-reconcile", 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.
skills CLI
$ npx skills add Abilityai/cornelius --skill ref-reconcile -a cursor
Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.
Install the "ref-reconcile" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/ref-reconcile into .cursor/skills/ref-reconcile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ref-reconcile", 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.
--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
skills CLI
$ npx skills add Abilityai/cornelius --skill ref-reconcile -a gemini-cli
Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.
Install the "ref-reconcile" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/ref-reconcile into .gemini/skills/ref-reconcile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ref-reconcile", 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.
Installs 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).
skills CLI
$ npx skills add Abilityai/cornelius --skill ref-reconcile -a github-copilot
Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.
Install the "ref-reconcile" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/ref-reconcile into .github/skills/ref-reconcile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ref-reconcile", 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.
skills CLI
$ npx skills add Abilityai/cornelius --skill ref-reconcile -a opencode
OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
Install the "ref-reconcile" agent skill from https://github.com/Abilityai/cornelius/tree/main/.claude/skills/ref-reconcile into .opencode/skills/ref-reconcile/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ref-reconcile", 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
ref-reconcile
GitHub stars
109
Token cost
~2k tokens
SKILL.md length
746 words
Files
1
Skills in repo
56
Repo updated
First seen
Licence
MIT
At a glance
Keep a reference scope canonical and internally consistent. An agent skill from Abilityai/cornelius.
Works in 3 steps: SENSE (auto) → FIX (every write gated) → Reindex after any write
SKILL.md covers Composes, State Dependencies, Process and Verification, plus 1 more section
Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Ref Reconcile is an agent skill from Abilityai/cornelius. Keep a reference scope canonical and internally consistent. SENSE phase (auto) detects issues by invoking ref-audit; FIX phase (human-gated) resolves conflicts, repairs bidirectional link integrity (A cites [[B]] as client → does B back-link?), fixes/removes dangling links, and MERGES duplicate entities into one canonical note. Every write is human-gated — merge and delete stay human even in market/. Company is the default scope.
Its SKILL.md is about 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.
Read from SKILL.md and the folder at commit fd5e9a4. 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
Bash
Glob
Grep
AskUserQuestion
Skill
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 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
Ref Reconcile loads about 2k tokens when it runs. Until then it costs about 112 tokens; SKILL.md has 746 words of instructions outside code blocks.
Always· name and description, kept in context so the agent knows when to use it
~112
When it runs· the whole SKILL.md, loaded when a task matches
~2k
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
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.
Download SKILL.mdSave it as .claude/skills/ref-reconcile/SKILL.md (or your agent's skills folder).
name
ref-reconcile
description
Keep a reference scope canonical and internally consistent. SENSE phase (auto) detects issues by invoking ref-audit; FIX phase (human-gated) resolves conflicts, repairs bidirectional link integrity (A cites [[B]] as client → does B back-link?), fixes/removes dangling links, and MERGES duplicate entities into one canonical note. Every write is human-gated — merge and delete stay human even in market/. Company is the default scope.
1.2: Two fixes (validated 2026-07-08) — (a) the Step 1 relation-verb grep was case-sensitive with a 5-verb space-only set and matched 4 of ~28 real mutual…
Ref Reconcile
ℹ️ First, set expectations: before anything else, print one short line with this skill's version and its most recent change — the top entry of metadata.changelog above — e.g. ref-reconcile v1.0 — recent: initial version. Then proceed.
The act on top of ref-audit's report. ref-audit senses (read-only, autonomous); ref-reconcile fixes (human-gated). It protects the property that makes a reference scope a source of truth: one canonical note per real-world entity, with consistent bidirectional links and no dangling references.
Read first:resources/layered-brains/COMPANY-BRAIN-SCHEMA.md → "Canonical source of truth" + "The update model"; resources/layered-brains/REFERENCE-SCOPE-SCHEMA.md.
Composes
/ref-audit — the sense/detection pass (do not re-implement the scan here).
State Dependencies
Source
Location
Read
Write
Description
Company family
Brain/Company/*/*.md
✓
✓ (gated)
Entity notes reconciled
Audit report
resources/company-brain/reports/
✓
Consumed for detection
LBS index
resources/local-brain-search/data/
✓
✓ (reindex)
Rebuilt after any write
Process
Step 1 — SENSE (auto)
Detection has two sources — don't attribute to ref-audit what it doesn't compute:
From /ref-audit (structural, single-authored there — do not re-implement): dangling links, orphan entities, duplicate candidates, plus the frontmatter/provenance/enum/partition/as_of violations. Invoke it (or read its latest report).
In this skill (relational, needs the references-vs-relations judgment ref-audit deliberately avoids):
Conflicts — two notes asserting contradictory facts about one entity. Read the duplicate/near-name clusters ref-audit flags and compare their bodies.
Bidirectional-link asymmetry — but only for relation links, never reference links. A ## Relations entry that names a mutual tie (client: [[B]], Partner of: [[B]], Engagement: [[B]]) should reciprocate; a directional reference (built on [[Trinity]], competitor of [[X]], runs on [[Y]]) must not be forced to back-link. ref-audit reports the family CLEAN despite dozens of such directional links precisely because a blanket A→B⇒B→A check would be almost all false positives — so scope the reciprocity check to relation verbs only:bash
cd Brain/Company
# Mutual-tie relation links whose reciprocal is worth checking.
# Case-insensitive (-i) and connector-agnostic: matches BOTH "Client of [[..]]" and the colon
# form "client: [[..]]" the migrated notes actually use. The verb must LEAD the label (~12 chars)
# so prose bullets that merely contain a verb word (e.g. "…problem owners…") aren't flagged.
# No \b — BSD/macOS grep does not support it. TUNE THE VERB SET to the scope's real vocabulary.
grep -rniE '^- [^[]{0,12}(client|partner|vendor|counsel|owner|owns|engagement|collaborat|advis)[^[]*\[\[' */*.md
(The old space-only, capitalized set — Client of|Vendor to|… — matched only 4 of the family's ~28 mutual relations; the family is predominantly lowercase colon-form.) For each match, confirm the named target carries the reciprocal relation; if not, it's a repair candidate (Step 2).
Show full SKILL.md (428 more words)Show less
Step 2 — FIX (every write gated)
For each issue, propose the concrete edit, show the before/after, and get explicit approval (AskUserQuestion) before writing. CRM and market alike — merge/delete are human in every scope.
Bidirectional link repair: if A lists [[B]] (e.g. "Client of [[Ability AI]]") but B has no reciprocal relation, propose adding the back-link to B's ## Relations. Append a ## Change Log line on the edited note.
Dangling link: the target note doesn't exist → propose either (a) correct the link to the real entity, or (b) remove it, or (c) create the missing entity via /ref-ingest. Ask which.
Conflict: two notes assert contradictory facts about one entity → present both, ask which is current; the loser is either corrected (overwrite via the note body) or, if a validity window closed, handed to /ref-supersede.
Duplicate MERGE (the canonical act): two notes are the same real-world entity →
choose the canonical note (usually the better-linked / more complete),
fold the duplicate's unique facts + [[relations]] into it,
repoint every inbound link:grep -rl "\[\[<Duplicate Name>\]\]" Brain/Company/ → rewrite each to [[<Canonical Name>]],
remove the duplicate note only after step 3 leaves zero inbound [[links]] to it; if history or lingering references matter, keep a status: superseded + superseded_by: [[Canonical]] tombstone instead (which keeps the name resolvable),
append a ## Change Log line on the canonical note noting the merge + source — name the merged duplicate as plain text, NOT a [[wiki-link]]. A wiki-link to a hard-removed note is itself a fresh dangling link (it would fail the very check this skill enforces); a [[Duplicate]] reference may remain only on the tombstone path, where the note still exists.
Step 3 — Reindex after any write
bash
cd resources/local-brain-search && ./run_index.sh && ./run_daemon.sh restart # :7437, RESTART not reload
(No BDG bootstrap — reference notes are BDG-excluded.) If nothing was written (all issues declined), skip.
Verification
After a merge: grep -ril "<duplicate name>" Brain/Company/ returns only the canonical note (or a tombstone); no inbound [[link]] still points at the removed duplicate — including the canonical note's own Change Log line (reference the merged name as plain text, so the merge does not re-introduce the dangling link it just resolved).
After a back-link repair: the reciprocal relation exists on both notes.
Re-running /ref-audit shows the fixed issues cleared and no new dangling links introduced by the edits.
Every edited note carries a new ## Change Log line.
This skill must refuse to
Write any fix without explicit human approval of that specific edit (sense is auto; every fix is gated).
Merge/delete on a hunch — a merge requires confirmed same-entity identity.
Set any provenance but reference, or crystallize/lifecycle-classify/promote a record into a cognitive insight.
Ref Reconcile 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.
A skill your agent uses when a Head of People Ops, BizOps lead, or Internal Communications owner needs to draft and sequence an internal-only change-management communication — a re-org announcement…
A set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill…
Standard format for capturing and documenting insights in the knowledge base.
109 GitHub stars~616 tokensUpdated 15 days ago
Auto-check passed
Questions about Ref Reconcile
What does Ref Reconcile do?
Keep a reference scope canonical and internally consistent. An agent skill from Abilityai/cornelius. Ref Reconcile is an agent skill from Abilityai/cornelius. Keep a reference scope canonical and internally consistent.
How do I install Ref Reconcile in Claude Code?
Run `npx skills add Abilityai/cornelius --skill ref-reconcile -a claude-code`. Or copy the skill folder (.claude/skills/ref-reconcile in Abilityai/cornelius) into .claude/skills/ref-reconcile in your project. Claude Code loads it when a task matches its description.
How do I install Ref Reconcile in Codex?
Run `npx skills add Abilityai/cornelius --skill ref-reconcile -a codex`. Or copy the skill folder (.claude/skills/ref-reconcile in Abilityai/cornelius) into .agents/skills/ref-reconcile in your project. Codex loads it when a task matches its description.
Can I use Ref Reconcile 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 ref-reconcile -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ref-reconcile, .gemini/skills/ref-reconcile, .github/skills/ref-reconcile and .opencode/skills/ref-reconcile in your project.
What does Ref Reconcile need to run?
SKILL.md names no scripts, command-line tools or credentials: Ref Reconcile is instructions for the agent only. Its frontmatter pre-approves these tools: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion, Skill.
Does Ref Reconcile 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 Ref Reconcile 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 Ref Reconcile use?
Ref Reconcile 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 Ref Reconcile use?
About 2k tokens (SKILL.md is roughly 8.1k 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 Ref Reconcile?
Skills that share tags, products or a category with Ref Reconcile: Consistency Internal (hashgraph-online/awesome-codex-plugins, 1.2k stars), Internal Comms (alirezarezvani/claude-skills, 28k stars), Internal Communication (sickn33/agentic-awesome-skills, 47k stars) and Internal Communications Writer (anthropics/skills, 180k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Ref Reconcile?
Abilityai (a GitHub organization) maintains it in Abilityai/cornelius, which has 109 GitHub stars. The repository holds 56 skills in this directory. The repository was last updated on September 22, 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.
