orchestrate

A command that assigns a development task to specialist agents through a staged PACT workflow. The workflow breaks work into preparation, design, coding, and testing phases, with additional phases when the task needs to be split into smaller parts.

In plain words
What is it for?
Use it to delegate a feature or other development task across preparation, architecture, implementation, decomposition when needed, consolidation, and testing.
Why use it?
It provides a visible order for complex work and makes later phases wait for the earlier work they depend on.

Command

Part of the pact-plugin plugin — 21 skills, 13 commands, 13 agents, 11 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/synaptic-labs-ai/pact-plugin/orchestrate
Clone the repo
git clone --depth 1 https://github.com/Synaptic-Labs-AI/PACT-Plugin

Or install pact-plugin, the plugin that ships this one along with the rest of its 21 skills, 13 commands, 13 agents, 11 hooks.

Per session 8 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 15,560 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 $0.00008 $0.15560
Opus 5 $0.00004 $0.07780
Sonnet 5 $0.00002 $0.03112
Haiku 4.5 $0.00001 $0.01556

Measured 3d ago against content hash bda33b402c2e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 3d 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.

pact-plugin/commands/orchestrate.md · 979 lines

How it starts

The opening of the file, as written. The whole thing — 979 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Orchestrate specialist PACT agents through the PACT workflow to address: $ARGUMENTS


Task Hierarchy

Create the full Task hierarchy upfront for workflow visibility:

1. `TaskCreate`: Feature task "{verb} {feature}"
2. `TaskCreate`: Phase tasks (all upfront):
   - "PREPARE: {feature-slug}"
   - "ARCHITECT: {feature-slug}"
   - "CODE: {feature-slug}"
   - "TEST: {feature-slug}"
3. `TaskUpdate`: Set phase-to-phase blockedBy chain:
   - ARCHITECT blockedBy PREPARE
   - CODE blockedBy ARCHITECT
   - TEST blockedBy CODE
4. `TaskUpdate`: Feature task status = "in_progress"

Scoped PACT phases: When decomposition fires after PREPARE, the standard ARCHITECT and CODE phases are skipped (decomposition_active) and replaced by scoped phases. Create retroactively (detection occurs after PREPARE):

  • "ATOMIZE: {feature-slug}" with blockedBy = [PREPARE task ID]
  • "CONSOLIDATE: {feature-slug}" with blockedBy = [all scope task IDs]
  • Add CONSOLIDATE to TEST's blockedBy via addBlockedBy = [CONSOLIDATE task ID] (the original CODE dependency auto-resolves when CODE is marked completed/skipped)

The scoped flow is: Prepare → Atomize → Consolidate → Test (same PACT acronym, scoped meanings).

For each phase execution:

a. `TaskUpdate`: phase status = "in_progress"
b. Analyze work needed (QDCL for CODE)
c. `TaskCreate`: agent task(s) as children of phase
d. `TaskUpdate`: agent tasks owner = "{agent-name}"
e. `TaskUpdate`: next phase addBlockedBy = [agent IDs]
f. Spawn teammates with the canonical dispatch form (shown at each specific phase below)
g. Monitor via `SendMessage` (completion summaries) and `TaskList` until agents complete
h. `TaskUpdate`: phase status = "completed" (agents self-manage their task status)

The canonical Agent() dispatch form, referenced by every phase below:

Agent(
  name="{teammate-name}",
  team_name="{team_name}",
  subagent_type="pact-{teammate-type}",
  description="Spawn {teammate-name} specialist",
  prompt="YOUR PACT ROLE: teammate ({teammate-name}).\n\nYou are joining team {team_name}. As your FIRST action, Invoke Skill(\"PACT:pact-team-registration\") to record your identity. Then check `TaskList` for tasks assigned to you."
)

Read the full file on GitHub · 979 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. 3d ago First seen · 979 lines · 8 tokens per session scan A bda33b402c2e

Subscribe to this mod's changes

orchestrate is a command published in the GitHub repository Synaptic-Labs-AI/PACT-Plugin (71 stars, last pushed 3d ago), licensed MIT. It adds 8 tokens to every session and 15,560 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 2026-08-30.