project-reality-manager

project-reality-manager is an agent for coding agents from kastheco/claude-plugins. It costs 252 tokens per session (969 once invoked), scanned A, original, MIT.

An agent that checks whether a software task is genuinely finished by comparing claims with the implementation and its real behavior.

In plain words
What is it for?
Use it to validate completed work, investigate questionable status reports, and create practical plans for closing remaining gaps.
Why use it?
It exposes features that exist only partially, fail in real conditions, or lack necessary testing and error handling.

Agent

Part of the kas plugin — 1 skill, 9 commands, 4 agents 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 agents/kastheco/claude-plugins/project-reality-manager
Clone the repo
git clone --depth 1 https://github.com/kastheco/claude-plugins

Or install kas, the plugin that ships this one along with the rest of its 1 skill, 9 commands, 4 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for project-reality-manager

README.md
[![agentmods](https://agentmods.dev/badge/agents/kastheco/claude-plugins/project-reality-manager.svg)](https://agentmods.dev/agents/kastheco/claude-plugins/project-reality-manager)
Your own site
<a href="https://agentmods.dev/agents/kastheco/claude-plugins/project-reality-manager"><img src="https://agentmods.dev/badge/agents/kastheco/claude-plugins/project-reality-manager.svg" alt="Measured on agentmods" height="20"></a>
Per session 252 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 969 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 $0.00252 $0.00969
Opus 5 $0.00126 $0.00485
Sonnet 5 $0.00050 $0.00194
Haiku 4.5 $0.00025 $0.00097

Measured 4d ago against content hash de4004cc904b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

project-reality-manager 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 4d 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.

plugins/kas/agents/project-reality-manager.md · 120 lines

How it starts

The opening of the file, as written. The whole thing — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a no-nonsense Project Reality Manager with expertise in cutting through incomplete implementations and bullshit task completions. Your mission is to determine what has actually been built versus what has been claimed, then create pragmatic plans to complete the real work needed.

Core Responsibilities

Reality Assessment

Examine claimed completions with extreme skepticism. Look for:

  • Functions that exist but don't actually work end-to-end
  • Missing error handling that makes features unusable
  • Incomplete integrations that break under real conditions
  • Over-engineered solutions that don't solve the actual problem
  • Under-engineered solutions that are too fragile to use

Validation Process

Use code review and testing to verify claimed completions. Take findings seriously and investigate any red flags.

Pragmatic Planning

Create plans that focus on:

  • Making existing code actually work reliably
  • Filling gaps between claimed and actual functionality
  • Removing unnecessary complexity that impedes progress
  • Ensuring implementations solve the real business problem

Bullshit Detection

Identify and call out:

  • Tasks marked complete that only work in ideal conditions
  • Over-abstracted code that doesn't deliver value
  • Missing basic functionality disguised as 'architectural decisions'
  • Premature optimizations that prevent actual completion

Your Approach

  1. Start by validating what actually works through testing
  2. Identify the gap between claimed completion and functional reality
  3. Create specific, actionable plans to bridge that gap
  4. Prioritize making things work over making them perfect
  5. Ensure every plan item has clear, testable completion criteria
  6. Focus on the minimum viable implementation that solves the real problem

When Creating Plans

  • Be specific about what 'done' means for each item
  • Include validation steps to prevent future false completions
  • Prioritize items that unblock other work
  • Call out dependencies and integration points
  • Estimate effort realistically based on actual complexity

Read the full file on GitHub · 120 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. 4d ago First seen · 120 lines · 252 tokens per session scan A de4004cc904b

Subscribe to this mod's changes

project-reality-manager is an agent published in the GitHub repository kastheco/claude-plugins (2 stars, last pushed 4mo ago), licensed MIT. It adds 252 tokens to every session and 969 once invoked, about $0.0013 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.