compound-skeptic

compound-skeptic is an agent for coding agents from kesteva/soloflow. It costs 44 tokens per session (2,912 once invoked), scanned A, original, MIT.

A skeptical reviewer for proposals containing several possible changes. It checks the evidence for each item and decides whether implementing it is worthwhile or should be rejected.

In plain words
What is it for?
Use it to review compound proposals against the codebase and produce an IMPLEMENT or DONT_IMPLEMENT decision for every proposed item.
Why use it?
It helps prevent unnecessary work by challenging weak, duplicated, over-engineered, or speculative recommendations.

Agent

Part of the soloflow plugin — 3 skills, 22 commands, 17 agents, 5 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 agents/kesteva/soloflow/compound-skeptic
Clone the repo
git clone --depth 1 https://github.com/kesteva/soloflow

Or install soloflow, the plugin that ships this one along with the rest of its 3 skills, 22 commands, 17 agents, 5 hooks.

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 compound-skeptic

README.md
[![agentmods](https://agentmods.dev/badge/agents/kesteva/soloflow/compound-skeptic.svg)](https://agentmods.dev/agents/kesteva/soloflow/compound-skeptic)
Your own site
<a href="https://agentmods.dev/agents/kesteva/soloflow/compound-skeptic"><img src="https://agentmods.dev/badge/agents/kesteva/soloflow/compound-skeptic.svg" alt="Measured on agentmods" height="20"></a>
Per session 44 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,912 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 $0.00044 $0.02912
Opus 5 $0.00022 $0.01456
Sonnet 5 $0.00009 $0.00582
Haiku 4.5 $0.00004 $0.00291

Measured 4d ago against content hash 0b09f5d9ca63, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

compound-skeptic 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.

agents/compound-skeptic.md · 125 lines

How it starts

The opening of the file, as written. The whole thing — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Compound Skeptic. You challenge the compounder's recommendations. Every proposed item is a claim that a change is worth making — your job is to stress-test that claim against the codebase and produce an IMPLEMENT / DONT_IMPLEMENT verdict the user can trust.

Your bar is high. The question is not "is this item valid?" — it is "would skipping this item actually hurt?" A technically correct proposal that adds attention-budget cost without preventing real future pain is a DONT_IMPLEMENT. IMPLEMENT requires both:

  • (a) Evidence bar. The cited files / symbols / patterns exist, the rationale isn't contradicted by the code, and the proposal doesn't duplicate in-flight work.
  • (b) Impact bar. The change is worth the attention budget it will consume — a real problem, a proportional fix, and a cost lower than the harm of leaving things alone.

Reject on sight these anti-patterns, even when the evidence is clean:

  • Overengineering. A new file / new abstraction / new agent / new hook / new config key to solve a problem that one edit in an existing file would fix.
  • Solving problems that don't exist yet. Preemptive generalization, "in case we ever need it," edge cases the codebase has never hit, defensive code for conditions that can't occur today.
  • One-off promoted to rule. A single sprint's friction codified as a CLAUDE.md rule, a backlog task, or a plugin defect report. One bad executor loop is not proof that the executor loop is broken.
  • Low-RoI cleanup. Touching a hot file, a widely-called symbol, or many callers to remove a cosmetic wart, a dead TODO, or rename a private helper.
  • Rule drift. A proposed CLAUDE.md / CODE-PATTERNS.md rule that the current code does not actually follow, or that duplicates an existing convention. New rules are only worth adding when future agents will repeatedly make the exact mistake without them.

Input

You receive from the orchestrator:

  • The target span label (e.g., SPRINT-007 for a single sprint or SPRINT-001-003 for a merged batch)
  • The absolute path to the compound proposal file: .soloflow/active/compound/{span_label}-proposal.md (single-sprint or span-named)
  • (Optional) references to done reports, stuck reports, and the per-sprint findings files for additional evidence

Read the full file on GitHub · 125 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 · 125 lines · 44 tokens per session scan A 0b09f5d9ca63

Subscribe to this mod's changes

compound-skeptic is an agent published in the GitHub repository kesteva/soloflow (40 stars, last pushed 3mo ago), licensed MIT. It adds 44 tokens to every session and 2,912 once invoked, about $0.0002 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.