start

A command for starting a feature in an isolated Git worktree, with a written plan and a first failing test before production code.

In plain words
What is it for?
Creating a feature worktree, optionally linking a Jira ticket and branch name, drafting the plan, and adding the initial failing test.
Why use it?
It keeps feature work separate and applies test-driven development (TDD), a method that starts by describing the expected behaviour in a test.

Command

Part of the orc plugin — 13 skills, 30 commands, 14 agents, 5 hooks shipped together

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 commands/higoralves/orc/start
Clone the repo
git clone --depth 1 https://github.com/HigorAlves/orc

Or install orc, the plugin that ships this one along with the rest of its 13 skills, 30 commands, 14 agents, 5 hooks.

Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,052 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00043 $0.02052
Opus 5 $0.00022 $0.01026
Sonnet 5 $0.00009 $0.00410
Haiku 4.5 $0.00004 $0.00205

Measured 2d ago against content hash 470e867ac21b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

start scanned grade A with 0 findings 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

orc/commands/start.md · 128 lines

How it starts

The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/orc:start

Kick off a new feature with isolation, a written plan, and the first failing test — all before production code is touched. The TDD red light is the entry gate.

Arguments

  • --auto[=guided|full] — autopilot level for this run (overrides interaction_policy; taxonomy in orc:using-orc). Soft-inward gates consult the resolved policy — guided auto-advances mechanical confirms with a printed one-liner; full pre-approves them from settled decisions (orc:state-protocol), stopping only on escalation-only conditions. Hard-outward gates are unaffected at every level.

  • --worktree <path> — optional explicit worktree directory. If omitted, the worktree skill picks one safely.

  • --jira <KEY> — optional. Link a Jira ticket key (e.g. JRA-123) to this session. Two effects: (1) Phase 1 fetches the ticket summary via acli jira workitem view <KEY> --fields "summary" --json, slugifies it, and offers feat/<KEY>-<slug> as the suggested branch name to orc:using-git-worktrees; (2) the flag is forwarded to /orc:plan in Phase 2, suppressing the Phase 1 link prompt and writing jiraTicket: <KEY> into the session's .orc/ state. Validate against ^[A-Z][A-Z0-9_]*-\d+$.

Workflow

Phase 0 — Detect context

!orc-workspace-detect --banner

Context is injected above (ORC_* vars are exported for any Bash you run — do not re-run detection).

In workspace mode, resolve targetRepos from --repos/--repo/--all-repos/--this-repo or via AskUserQuestion. Iron rule: no silent broadcast.

Phase 1 — Worktree(s)

Invoke orc:using-git-worktrees. Create an isolated worktree off the current default branch (typically main). Switch the session to that worktree. The PreToolUse hook will refuse subsequent commits to main/master/develop, so you must be on a feature branch from this point.

If --jira <KEY> was passed: invoke orc:jira-cli and run acli jira workitem view "$KEY" --fields "summary" --json | jq -r '.fields.summary'. Slugify (lowercase, replace non-[a-z0-9-] with -, collapse repeats, trim leading/trailing -). Suggest feat/<KEY>-<slug> as the branch name to orc:using-git-worktrees. The user can accept or override.

Read the full file on GitHub · 128 lines

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 · 128 lines · 43 tokens per session scan A 470e867ac21b

Subscribe to this mod's changes

start is a command published in the GitHub repository HigorAlves/orc (6 stars, last pushed 6d ago), licensed MIT. It adds 43 tokens to every session and 2,052 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.