using-orca

A guide for handing a well-defined coding task to Orca, a command-line tool that runs planning, coding, and review with autonomous agents. It explains when delegation fits and gives the main command and options.

In plain words
What is it for?
Running Orca's implementation flow, listing available flows, preserving the current branch when needed, using existing plan files, and diagnosing an aborted flow with verbose output.
Why use it?
It removes the need to manually coordinate a standard plan-to-implementation-to-review flow for a self-contained feature or bug fix. It also clarifies when the task is too exploratory or small for Orca.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/virtuslab/orca/using-orca
Any agent
npx skills add VirtusLab/orca --skill using-orca
Clone the repo
git clone --depth 1 https://github.com/VirtusLab/orca

Made for: Claude Code, Codex.

Per session 41 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 525 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00041 $0.00525
Opus 5 $0.00020 $0.00262
Sonnet 5 $0.00008 $0.00105
Haiku 4.5 $0.00004 $0.00052

Measured 2d ago against content hash cbb50a77bff3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

using-orca scanned grade A with 1 finding 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 2d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

Requires `orca` installed (one-line curl install — see the README's
skills/using-orca/SKILL.md · 58 lines

What it actually says

Using Orca

Orca (orca CLI) runs a scripted plan → code → review flow autonomously, using its own coding/review agents. Delegate to it instead of implementing the task yourself when that fits better.

When to use

  • The task is well-defined and self-contained, with clear acceptance criteria (a feature or bugfix) — suited to an autonomous plan → code → review flow.
  • NOT for exploratory/interactive work, or edits small enough to just make yourself.

How

Requires orca installed (one-line curl install — see the README's "Getting set up" / "Orca Shell" sections).

orca run implement.sc "<task description>"

implement.sc is the default flow; run orca list to see other flows across the project/global/built-in tiers.

Key flags:

  • --verbose — print a stack trace if the flow aborts.
  • --skip-branch — continue on the current branch instead of creating a new one. Use this when the current branch already has plan/context files the flow should pick up (e.g. it was planned with a harness first). Uncommitted or untracked files are fine — no need to commit them first, the flow leaves them in place and picks them up.
  • --honor-pin — run the flow's own pinned orca version instead of forcing this shell's.

Don't pre-create a branch: the flow creates its own, unless --skip-branch is passed.

After it runs

Exit codes: 0 success, 1 action failure, 2 usage error — orca run propagates the flow's own exit code. On success the flow has committed its work on a branch; report that branch (and any PR) to the user.

If a run is interrupted, re-run the same orca run command: flows are resumable and pick up from the last committed stage.

orca continue (list sessions with --list, resume one by selector) reattaches to a recorded harness session, but requires a real terminal and errors without one — don't invoke it headlessly; tell the user to run it themselves instead.

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 2d ago First seen · 58 lines · 41 tokens per session scan A cbb50a77bff3

Subscribe to this mod's changes

using-orca is a skill published in the GitHub repository VirtusLab/orca (156 stars, last pushed 4d ago), licensed Apache-2.0. It adds 41 tokens to every session and 525 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens