New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Omni

by fajarhide in fajarhide/omni

A skill your agent uses when installing, verifying or configuring OMNI, when command output carries an [OMNI: ...] marker, or when output looks shorter than expected.

Apache-2.0Auto-check passedAI & LLM Engineering

Install Omni

skills CLI
$ npx skills add fajarhide/omni --skill omni -a claude-code

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

GitHub CLI
$ gh skill install fajarhide/omni omni --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/fajarhide/omni.git skills-src && mkdir -p .claude/skills && cp -r skills-src/plugins/claude-code/skills/omni .claude/skills/omni && 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
omni
GitHub stars
374
Token cost
~1.1k tokens
SKILL.md length
630 words
Files
1
Skills in repo
1
Repo updated
First seen
Licence
Apache-2.0

At a glance

A skill your agent uses when installing, verifying or configuring OMNI, when command output carries an [OMNI: ...] marker, or when output looks shorter than expected.

  • Configuring OMNI
  • SKILL.md covers Install, Verify, Reading a marker and When output looks wrong rather…
  • Calls brew and claude
  • Command output carries an [OMNI: ...] marker

What it does

Omni is an agent skill from fajarhide/omni. Use when installing, verifying or configuring OMNI, when command output carries an [OMNI: ...] marker, or when output looks shorter than expected. OMNI is a local hook that shortens shell output before the model reads it and hands back a handle for what it cut.

Its SKILL.md is about 1.1k 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 AI & LLM Engineering, covering LLM cost and token optimization. It works with Model Context Protocol, Homebrew and Rust. The repository describes itself as: Your agent pays twice for output it has already seen. OMNI returns a handle instead: 97.2% off a file read twice. Nothing deleted, nothing invented. The licence is Apache-2.0.

When your agent uses it

  • Configuring OMNI
  • Command output carries an [OMNI: ...] marker
  • Output looks shorter than expected

Example prompts

  • “/omni”

What it can do on your machine

Read from SKILL.md and the folder at commit d351899. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • brew
    • claude

    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

Omni loads about 1.1k tokens when it runs. Until then it costs about 67 tokens; SKILL.md has 630 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~67
When it runs · the whole SKILL.md, loaded when a task matches
~1.1k

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 passed

The automated check found no risky patterns in 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.

SKILL.md

The full file from fajarhide/omni at commit d351899, republished under its Apache-2.0 licence (© fajarhide). 630 words, ~1,109 tokens.

Download SKILL.mdSave it as .claude/skills/omni/SKILL.md (or your agent's skills folder).
name
omni
description
Use when installing, verifying or configuring OMNI, when command output carries an `[OMNI: ...]` marker, or when output looks shorter than expected. OMNI is a local hook that shortens shell output before the model reads it and hands back a handle for what it cut.

OMNI

OMNI runs on the machine, as a hook the agent host calls. It rewrites the output of a shell tool call before the model reads it, and every cut it makes leaves a marker and a handle to fetch the bytes back. No account, no telemetry, nothing about the machine leaves it.

Install

brew install fajarhide/tap/omni
omni init

Without Homebrew, take the release archive for the platform from https://github.com/fajarhide/omni/releases, pull the line for that archive out of the SHA256SUMS published beside it into a file of its own, and verify it with sha256sum -c <that file>, or shasum -a 256 -c <that file> on macOS. Then untar and put the binary on the path. Two things decide whether the check is worth anything. The manifest line has to reach the tool as a file rather than a pipe, because a grep that matches nothing sends an empty stream and macOS reads that as a pass, so a mistyped archive name would install unverified. And the verification has to gate the untar with &&, because a shell running pasted lines carries on past a failed one.

Do not name a host you have not established you are running in. With no terminal to draw its menu on, which is the case when an agent runs it, omni init configures the host it is running inside and prints which one it picked. If it cannot name the host, a plain shell for instance, it stops and lists the flags rather than installing somewhere nobody asked for. Only then pick one: --claude, --cursor, --codex, --gemini and the rest under omni init --help. omni init --all takes every supported host and also writes a .vscode/mcp.json in the working directory.

The hooks land in that host's own configuration, and the host has to restart before they run, so say that rather than reporting the install as active.

Verify

omni doctor

A healthy install names the binary, the config directory, the database, and one line per host:

  Binary:         omni v0.7.5 [LATEST]
  Config dir:     ~/.omni/ [OK]
  Database:       ~/.omni/omni.db (0 distillations, 0 sessions) [OK]

 Agent Integrations:
  Claude Code: Full, 5 checks [OK]

Full is the line to check: it means this host applies the rewrite, so the model reads shortened output. MCP-only means memory and session state and no shell distillation. omni doctor --fix repairs what it can, and reports what it could not rather than claiming success.

Show full SKILL.md (260 more words)Show less

Reading a marker

A marker means bytes were cut and can be fetched back. Do not treat the shortened output as the whole output, and do not write a file back from output that carries one.

markerwhat happened
[OMNI: 40 lines omitted, omni retrieve <handle> for full output]the distiller kept the signal and archived the rest
[OMNI: 40 lines already shown, omni retrieve <handle>]those lines are in this session's context already
[OMNI: identical to the 40 lines already shown, omni retrieve <handle>]the whole re-run matched an earlier one
[OMNI: 40 lines not shown here, omni retrieve <handle>]another session in this project saw them and this one never did, so retrieve before relying on them
[OMNI: identical to 40 lines from an earlier session, none shown here, omni retrieve <handle>]the whole reply matched an earlier session's, and none of it arrived here
[OMNI: 2 sensitive value(s) redacted]values under a key that names a credential

Pull the full bytes with the command the marker prints:

omni retrieve <handle>

The omni_retrieve MCP tool does the same without a shell. Reach for either whenever the answer depends on what was cut: a count, an exact line, or a file being rewritten from what was read.

When output looks wrong rather than short

Corrupted output, a success reported for something that failed, or a marker written into a file is a bug, not compression. Capture the command and its output and file it at https://github.com/fajarhide/omni/issues, then get a clean copy by running the command through omni exec with the pipeline off:

OMNI_PASSTHROUGH=1 omni exec <the command>

© fajarhide, Apache-2.0. 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 plugins/claude-code/skills/omni of fajarhide/omni.

Open the folder on GitHubat commit d351899

Compare with similar skills

Omni 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.

Omni compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Omni this skillfajarhide/omni374—~1.1kAutomated safety check: PassApache-2.0
Agents Best PracticesDenisSergeevitch/agents-best-practices2.4k—~7.4kAutomated safety check: PassMIT
Flowflow Spacesmirkobozzetto/flowflow171—~1kAutomated safety check: PassEUPL-1.2
Agentbro Releaseshirenchuang/agentbro203—~2.5kAutomated safety check: PassApache-2.0
CodexBar Usage Readersteipete/CodexBar22k—~320Automated safety check: PassMIT
Agenttrace Session Auditluoyuctl/agenttrace1401 repos~656Automated safety check: PassMIT

Similar skills

  • Agents Best Practices

    DenisSergeevitch/agents-best-practices

    A skill your agent uses when designing, generating an MVP blueprint for, auditing, troubleshooting, refactoring, or explaining an agentic harness for any domain.

    2.4k GitHub stars~7.4k tokensUpdated 3 days ago
    AI & LLM EngineeringAuto-check passed

  • Flowflow Spaces

    mirkobozzetto/flowflow

    Use FlowFlow spaces safely through their scoped MCP server. An agent skill from mirkobozzetto/flowflow.

    171 GitHub stars~1k tokensUpdated today
    AI & LLM EngineeringAuto-check passed

  • Agentbro Release

    shirenchuang/agentbro

    A skill your agent uses when releasing AgentBro from this repository: merging dev/main, bumping versions, updating release notes, tagging, pushing, monitoring GitHub Actions, Homebrew cask…

    203 GitHub stars~2.5k tokensUpdated 7 days ago
    DevelopmentAuto-check passed

  • CodexBar Usage Reader

    steipete/CodexBar

    CodexBar read. Provider usage, limits, credits, config health. JSON. No writes.

    22k GitHub stars~320 tokensUpdated today
    AI & LLM EngineeringAuto-check passed

  • Agenttrace Session Audit

    luoyuctl/agenttrace

    Audit local AI coding-agent sessions with agenttrace. An agent skill from luoyuctl/agenttrace.

    140 GitHub starsUsed in 1 repo~656 tokens
    AI & LLM EngineeringAuto-check passed

  • Evaluating Bitrouter Routes

    bitrouter/bitrouter

    A skill your agent uses when evaluating BitRouter route decisions or Eval Exchange subjects with task-native verifiers, human reviewers, private enterprise evaluators, agentic judges, or genuinely…

    235 GitHub stars~1.2k tokensUpdated today
    AI & LLM EngineeringAuto-check passed

Works with

Categories

Questions about Omni

What does Omni do?

A skill your agent uses when installing, verifying or configuring OMNI, when command output carries an [OMNI: ...] marker, or when output looks shorter than expected. Omni is an agent skill from fajarhide/omni.] marker, or when output looks shorter than expected.

When should I use Omni?

Omni fits situations like: configuring OMNI; command output carries an [OMNI: ...] marker; output looks shorter than expected.

How do I install Omni in Claude Code?

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

How do I install Omni in Codex?

Run `npx skills add fajarhide/omni --skill omni -a codex`. Or copy the skill folder (plugins/claude-code/skills/omni in fajarhide/omni) into .agents/skills/omni in your project. Codex loads it when a task matches its description.

Can I use Omni 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 fajarhide/omni --skill omni -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/omni, .gemini/skills/omni, .github/skills/omni and .opencode/skills/omni in your project.

What does Omni need to run?

Going by SKILL.md and its folder, Omni needs the command-line tools its instructions call (brew and claude).

Does Omni 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 Omni safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Omni use?

Omni is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Omni use?

About 1.1k tokens (SKILL.md is roughly 4.4k 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 Omni?

Skills that share tags, products or a category with Omni: Agents Best Practices (DenisSergeevitch/agents-best-practices, 2.4k stars), Flowflow Spaces (mirkobozzetto/flowflow, 171 stars), Agentbro Release (shirenchuang/agentbro, 203 stars) and CodexBar Usage Reader (steipete/CodexBar, 22k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Omni?

fajarhide (a GitHub user) maintains it in fajarhide/omni, which has 374 GitHub stars. The repository was last updated on October 6, 2026.

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