feature-orchestrate

An autonomous command that runs a prepared feature plan through multiple implementation phases using background agents.

In plain words
What is it for?
Use it to execute an existing feature plan to completion, recover its documented state, and accept phases by checking their required evidence.
Why use it?
It lets a planned feature progress without waiting for the user between phases while recording decisions and checking evidence.

Command

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/langerrr/zforge/feature-orchestrate
Clone the repo
git clone --depth 1 https://github.com/Langerrr/zforge
Per session 6 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,230 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 $0.00006 $0.01230
Opus 5 $0.00003 $0.00615
Sonnet 5 $0.00001 $0.00246
Haiku 4.5 $0.00001 $0.00123

Measured 2d ago against content hash 361d15f57aee, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

feature-orchestrate 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.

commands/feature-orchestrate.md · 75 lines

How it starts

The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/feature-orchestrate — Autonomous Feature Execution

Run a feature plan to completion without blocking on the user. Phases execute as background subagents; decisions are recorded for asynchronous review; the planner accepts each phase by re-running its evidence.

Load the feature-execution skill. It owns phase state, the spawn contract, recovery, scheduling, and acceptance. This command supplies the autonomous intent only.

Arguments

  • $1: Feature name (snake_case, or will be converted)

Pre-flight

  1. Convert the feature name to snake_case and resolve docs/{feature_name}/. If the directory is missing, report it and suggest /zforge:plan.
  2. Read 05_progress_overview.md, 01_context.md, 02_plan.md, and the Doc Map in 01_context.md for anything else binding.
  3. Read decision_review.md. If it does not exist, create it from ${CLAUDE_PLUGIN_ROOT}/templates/decision_review.md.
  4. Read session_log.md, creating it from ${CLAUDE_PLUGIN_ROOT}/templates/session_log.md if absent, and append a row for this session.
  5. Note any open standing flags in the overview. A flag opened by an earlier session is inherited by this one.
  6. Read 07_harness_conventions.md if the feature has one. It says how this project is run and observed, and it is what keeps the run from re-learning a fact an earlier phase already paid for.

If the feature predates v3 — phase files with no ## Evidence Required table — say so once, treat every phase's evidence as E0, and offer to backfill the Verification Matrix in 02_plan.md before executing. Executing without it is allowed; it just means acceptance can only check that commands run, not that they were the right class.

If any phase is REPORTED — an earlier session took its report and died before verifying it — acceptance is the first work of this session. Do not resume its agent.

Read the full file on GitHub · 75 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 · 75 lines · 6 tokens per session scan A 361d15f57aee

Subscribe to this mod's changes

feature-orchestrate is a command published in the GitHub repository Langerrr/zforge (10 stars, last pushed 2d ago), licensed MIT. It adds 6 tokens to every session and 1,230 once invoked, about $0.0000 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.