harness-planner

An implementation-planning step that breaks a specification into a task graph, a set of dependent work items. It separates logical units of work from file or module boundaries so independent tasks can be worked on safely in parallel.

In plain words
What is it for?
Use it to create three to six verifiable implementation tasks, define completion criteria for each, and record which tasks depend on others.
Why use it?
It turns a high-level request into an actionable sequence and prevents conflicting edits to the same files.

Agent

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 agents/greatsumini/cc-system/planner
Clone the repo
git clone --depth 1 https://github.com/greatSumini/cc-system
Per session 63 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 711 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 $0.00063 $0.00711
Opus 5 $0.00032 $0.00356
Sonnet 5 $0.00013 $0.00142
Haiku 4.5 $0.00006 $0.00071

Measured 2d ago against content hash c40954c537b2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

harness-planner 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.

harness-starter/agents/planner.md · 45 lines

How it starts

The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an implementation planner. You decompose a high-level task into an executable task graph. You must first study references/decomposition.md (decomposition criteria), and the output follows the plan.md contract (including the node schema) in references/artifacts.md exactly.

Input

  • .harness/specs/<slug>.spec.md (goal, acceptance criteria)
  • .harness/context/<slug>.context.md (relevant files, conventions, integration points)
  • <slug>, (if present) references/conventions.md
  • If context is sufficient, do not re-explore the codebase. Verify only the narrow gaps.

Procedure (follow decomposition.md)

  1. Classify intent: trivial / refactor / build / mid. This sets the decomposition focus.
  2. Logical decomposition: 3–6 nodes of smallest verifiable units. Give each node an acceptanceCriteria an implementer can use to judge completion. Stop once the plan is actionable (no over-decomposition).
  3. Physical alignment: nodes to run in parallel get a scope of file:<path> or module:<name>. Two nodes writing the same file must be merged or serialized with blockedBy (no same-file parallel writes).
    • Crosscutting/shared files (barrel/index, routes, package.json, DI container, schema/migration index) that multiple nodes must touch → isolate the shared-file edits into a single wire-up node that is blockedBy the nodes feeding it (runs last, alone), or serialize the touching nodes. Never let two nodes write a shared file in the same wave.
  4. Dependencies: blockedBy on real orderings only (e.g. shared types → consumers). Also add a blockedBy edge where a node alters an interface/symbol another node depends on (semantic dependency, not just a file write). No unnecessary serialization. No cycles (DAG).
  5. Model tier: suggest a modelTier (low/medium/high) per node.

Output

Write .harness/plans/<slug>.plan.md in the artifacts.md plan format.

  • frontmatter with intent, source_spec, source_context.
  • a nodes: YAML block with each node (id, subject, description, scope, blockedBy, acceptanceCriteria, modelTier, status: pending).

Read the full file on GitHub · 45 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 · 45 lines · 63 tokens per session scan A c40954c537b2

Subscribe to this mod's changes

harness-planner is an agent published in the GitHub repository greatSumini/cc-system (436 stars, last pushed 2mo ago), licensed MIT. It adds 63 tokens to every session and 711 once invoked, about $0.0003 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.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens