rePACT

A recursive command that runs a smaller Prepare, Architect, Code, and Test cycle for a complex sub-task inside a larger feature. It keeps the smaller work connected to the parent task.

In plain words
What is it for?
Use it to plan, design, implement, and test a complex sub-feature through nested tasks and specialist teammates.
Why use it?
It provides structure for difficult pieces of work without separating them from the main feature’s task tracking.

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/synaptic-labs-ai/pact-plugin/repact
Clone the repo
git clone --depth 1 https://github.com/Synaptic-Labs-AI/PACT-Plugin
Per session 10 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,328 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 $0.00010 $0.05328
Opus 5 $0.00005 $0.02664
Sonnet 5 $0.00002 $0.01066
Haiku 4.5 $0.00001 $0.00533

Measured 2d ago against content hash 7e55649fedf1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

rePACT 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.

pact-plugin/commands/rePACT.md · 460 lines

How it starts

The opening of the file, as written. The whole thing — 460 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Run a recursive PACT cycle for this sub-task: $ARGUMENTS

This command initiates a nested P→A→C→T cycle for a sub-task that is too complex for simple delegation but should remain part of the current feature work.

Team behavior: rePACT spawns sub-scope teammates into the session team ({team_name}). The platform provisions the team; rePACT does not create a new one. Use scope-prefixed names (e.g., backend-coder-auth-scope) to distinguish sub-scope teammates from parent-scope teammates.


Task Hierarchy

Create a nested Task hierarchy as a child of the current context:

1. `TaskCreate`: Sub-feature task "{verb} {sub-feature}" (child of parent context) — stamp metadata.variety.total (4-16)
2. `TaskCreate`: Nested phase tasks:
   - "PREPARE: {sub-feature-slug}"
   - "ARCHITECT: {sub-feature-slug}"
   - "CODE: {sub-feature-slug}"
   - "TEST: {sub-feature-slug}"
3. `TaskUpdate`: Set dependencies:
   - Phase-to-phase blockedBy chain (same as orchestrate)
   - Parent task addBlockedBy = [sub-feature task]
4. `TaskUpdate`: Sub-feature task status = "in_progress"
5. Execute nested P→A→C→T cycle (same per-phase lifecycle as orchestrate: create phase task → `in_progress` → dispatch agents → agent tasks `in_progress` → `completed` → phase `completed`)
6. On completion: Parent task unblocked

Example structure (standard):

[Feature] "Implement user auth" (parent, blockedBy: sub-feature)
└── [Sub-Feature] "Implement OAuth2 token refresh"
    ├── [Phase] "PREPARE: oauth2-token-refresh"
    ├── [Phase] "ARCHITECT: oauth2-token-refresh"
    ├── [Phase] "CODE: oauth2-token-refresh"
    └── [Phase] "TEST: oauth2-token-refresh"

Scope-aware naming (when scope contract is provided):

When a scope contract provides a scope_id, prefix all tasks with [scope:{scope_id}]:

[Feature] "Implement user auth" (parent, blockedBy: sub-feature)
└── [Sub-Feature] "[scope:backend-api] Implement backend API"
    ├── [Phase] "[scope:backend-api] PREPARE: backend-api"
    ├── [Phase] "[scope:backend-api] ARCHITECT: backend-api"
    ├── [Phase] "[scope:backend-api] CODE: backend-api"
    └── [Phase] "[scope:backend-api] TEST: backend-api"

Include scope_id in task metadata: { "scope_id": "backend-api" }. This enables the parent orchestrator to filter tasks by scope when aggregating results.


Read the full file on GitHub · 460 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 · 460 lines · 0 tokens per session scan A 7e55649fedf1

Subscribe to this mod's changes

rePACT is a command published in the GitHub repository Synaptic-Labs-AI/PACT-Plugin (71 stars, last pushed 2d ago), licensed MIT. It adds 10 tokens to every session and 5,328 once invoked, about $0.0001 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.