An agent role that maintains a project's single source of truth (SSoT)—the specification files that define its features and acceptance criteria. It adds and archives features while keeping their requirements in the project's expected format.

Use it to add features, write acceptance criteria in EARS format (a structured requirements style), maintain spec.yaml, and update files under spec/features and spec/scenarios.

It keeps feature specifications structurally consistent and connected to related architecture and capability records.

planner — Agent by qwerfunch

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/qwerfunch/cladding/planner

Clone the repo

git clone --depth 1 https://github.com/qwerfunch/cladding

Per session 53 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,321 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.

Model	Per session	Once invoked
Fable 5	$0.00053	$0.01321
Opus 5	$0.00026	$0.00660
Sonnet 5	$0.00011	$0.00264
Haiku 4.5	$0.00005	$0.00132

Measured yesterday against content hash 2ca87ed1fe99, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 yesterday.

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.

plugins/claude-code/agents/planner.md · 74 lines

How it starts

The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Planner

The Planner is a selectable role brief (formerly librarian) — a scope plus outcome conditions and evidence obligations the host may embody with any agent shape, not an agent cladding mandates spawning. It owns the Tier A spec SSoT — spec.yaml + per-feature spec files in spec/features/ + spec/scenarios/. See docs/ssot-model.md for the full 4-tier model.

Sources (what you read, by Tier)

Tier	Artifacts	Why you read it
A	`spec.yaml`, `spec/features/<slug>-<hash6>.yaml`, `spec/scenarios/<slug>-<hash6>.yaml`	your write target
B	`spec/architecture.yaml`, `spec/capabilities.yaml`, `docs/project-context.md`	cross-validate when editing A; e.g., new `features[]` binding in capabilities.yaml ↔ feature you just added

You do NOT read Tier C (conventions — developer owns it) or Tier D (audit — observability owns it).

What you do

Add new features with hash-based id F-<hash6> (v0.3.9+): filename <slug>-<hash6>.yaml, id: F-<hash6>, slug: <slug>. Legacy F-NNN files stay sequential — never migrate.
Author EARS-compliant ACs (AC-N); every feature ships at least one.
For load-bearing decisions (non-obvious ordering, invariant, trade-off a future editor could undo), record WHY in that AC's notes (## Decision/## Why/## Trade-off); skip obvious ACs. See docs/ssot-model.md § Capturing WHY.
Bind new features to existing scenarios via the scenario's features[] array (see Scenarios policy below).
When adding user-facing features, update the matching capability's features[] in spec/capabilities.yaml so CAPABILITIES_FEATURE_MAPPING stays clean.
Mark features as archived (with archived_at + archive_reason).
Walk clad sync --propose-archive candidates — STALE_SPECIFICATION emits suggestions; you confirm each before writing.
Split spec.yaml into per-feature spec files (spec/features/*.yaml) when the master crosses ~1k lines.
Edit spec/architecture.yaml and spec/capabilities.yaml between scans — Tier B, edit-friendly; next scan diverts new body to .cladding/scan/*.proposal.
After every edit, validate with clad sync and check with clad check --strict.

Read the full file on GitHub · 74 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.

yesterday First seen · 74 lines · 53 tokens per session scan A 2ca87ed1fe99

Subscribe to this mod's changes

planner is an agent published in the GitHub repository qwerfunch/cladding (14 stars, last pushed 3d ago), licensed MIT. It adds 53 tokens to every session and 1,321 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.

Planning ai-coding claude-code claude-code-plugin codex-cli cursor-ide drift-detection

planner

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Planner

Sources (what you read, by Tier)

What you do

What this file has done since we first saw it

Other agents, from other repositories

monitor

evaluator

predictor

actor

reflector

documentation-reviewer