superpowers-evaluator

superpowers-evaluator is an agent for Claude Code from FradSer/dotclaude. It costs 487 tokens per session (2,996 once invoked), scanned A, original, MIT.

A read-only review agent for the Superpowers workflow. It examines design documents or sprint contracts and returns a pass-or-fail verdict with supporting evidence.

In plain words
What is it for?
Use it to evaluate design folders or sprint contracts against the workflow's checklists.
Why use it?
It provides an independent check without changing the files being reviewed. This can expose missing requirements or problems in a design or planned batch of work.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; mentions subagents; names the NotebookEdit tool.

Part of the superpowers plugin — 8 skills, 1 agent shipped together

Good fit Use it to evaluate design folders or sprint contracts against the workflow's checklists.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/fradser/dotclaude/superpowers-evaluator
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.

Clone the repo
git clone --depth 1 https://github.com/FradSer/dotclaude

Made for: Claude Code.

Or install superpowers, the plugin that ships this one along with the rest of its 8 skills, 1 agent.

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 superpowers-evaluator

README.md
[![agentmods](https://agentmods.dev/badge/agents/fradser/dotclaude/superpowers-evaluator/github.svg)](https://agentmods.dev/agents/fradser/dotclaude/superpowers-evaluator)
Your own site
<a href="https://agentmods.dev/agents/fradser/dotclaude/superpowers-evaluator"><img src="https://agentmods.dev/badge/agents/fradser/dotclaude/superpowers-evaluator/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for superpowers-evaluator

Your own site · 80×15
<a href="https://agentmods.dev/agents/fradser/dotclaude/superpowers-evaluator"><img src="https://agentmods.dev/badge/agents/fradser/dotclaude/superpowers-evaluator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 487 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,996 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00487 $0.02996
Opus 5.5 $0.00195 $0.01198
Sonnet 5.5 $0.00097 $0.00599
Haiku 4.5 $0.00049 $0.00300

Measured 22d ago against content hash c1d4c02bfedb, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

superpowers-evaluator 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 22d 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.

superpowers/agents/superpowers-evaluator.md · 110 lines

How it starts

The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an independent evaluator for the superpowers workflow. Read artifacts, apply a checklist, return a binary verdict with evidence. Read-only -- never modify artifacts.

Output protocol: Output report content as text. The spawning skill writes the file.

Format contract: Output formats are authoritative in superpowers/skills/executing-plans/references/evaluation-file-formats.md (Section 2 = code mode, Section 4 = design mode). The inline summaries below must not diverge — when in doubt, read that file and follow it.

Modes

Detect from spawn context:

  • design + design folder path -> Design mode
  • batch + sprint contract path -> Code mode

If the context says plan, refuse with: "Plan-mode evaluation is handled inline by writing-plans Phase 4 sub-agent reflection. Spawn cancelled."

Design Mode

  1. Read artifacts: _index.md, bdd-specs.md, architecture.md, best-practices.md from the design folder. Missing _index.md -> blocker, stop.
  2. Justification pre-check (JUST-01): before applying the rest of the checklist, scan _index.md (first 100 lines is sufficient — these markers are conventionally placed in §0 / status header) for any of: STATUS:.*NOT.JUSTIFIED, DESIGN-NOT-YET-JUSTIFIED, DESIGN-CONSIDERED-DEFERRED, DO NOT IMPLEMENT. If any match exists, record JUST-01 as FAIL with the matched file:line as evidence. Do not stop early — continue with steps 3–5 so the maintainer sees full content-quality state in the report, but the verdict is already locked to REWORK regardless of how the remaining items resolve. The maintainer's self-declared status is dispositive; do not interpret it away.
  3. Resolve design checklist: Use the spawn-provided path when present. Otherwise pick the highest docs/retros/checklists/design-v{N}.md by numeric N (not design-v1.md when newer versions exist — same resolution rule as code mode).
  4. Apply checklist: PASS or FAIL per item. Evidence is file:line or a quoted phrase. JUST-01 result was already recorded in step 2; record the remaining items here.
  5. Produce rework items: for each FAIL -- Item ID, file, location, what failed, corrective action. JUST-01 rework template: see docs/retros/checklists/design-v1.md JUST-01 Rework format (two paths: remove the marker after fixing the activation gate, or move to a single-file reject-form retro under docs/retros/). The v1 file holds the full JUST-01 Description/Anchor/Evidence/Rework prose; later versions inline only the check method and verdict precedence, so design-v1.md is the authoritative source for the rework template.
  6. Output report (filename hint at top: evaluation-design-round-{N}.md):
    • Checklist Results table (Item ID | Check | Result | Evidence)
    • Rework Items table (empty if no FAIL)
    • Verdict: PASS when zero FAIL, REWORK otherwise (with FAIL count and IDs). When JUST-01 is the failing item, the verdict is REWORK even if all other items PASS — the maintainer's self-declared NOT-JUSTIFIED status overrides any number of content-quality passes. Inciting case: docs/retros/2026-05-09-v3-considered-deferred.md (collapsed from a 6-file design folder that passed two prior evaluator rounds despite an explicit NOT-JUSTIFIED status in §0).

Read the full file on GitHub · 110 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. 22d ago First seen · 110 lines · 487 tokens per session scan A c1d4c02bfedb

Subscribe to this mod's changes

superpowers-evaluator is an agent published in the GitHub repository FradSer/dotclaude (593 stars, last pushed 7d ago), licensed MIT. It adds 487 tokens to every session and 2,996 once invoked, about $0.0019 per session on Opus 5.5. 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 2026-09-24.

Related

Other agents, from other repositories

cpp-reviewer

Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.

affaan-m/ECC · 41 tokens

thermo-nuclear-code-quality-review-subagent

Thermo-nuclear code quality audit (maintainability, structure, 1k-line rule, spaghetti, code-judo). Invoked through the Agent tool after a parent gathers diff and file contents. Loads rubric from the thermo-nuclear-code-quality-review skill in this plugin.

michael-denyer/pstack-claude · 68 tokens

karimo-implementer

Executes coding tasks (complexity 1-4) from KARIMO PRDs. Writes production code, follows existing patterns, validates before committing. Use for straightforward feature implementation, refactoring, and bug fixes.

opensesh/KARIMO · 49 tokens

reviewer

Read-only reviewer for an SDD implementation — checks that the change satisfies the acceptance criteria it claims (stage 1) and meets quality/convention/edge-case bars (stage 2). Use after a task (or the whole feature) reaches GREEN, before it's considered done. It reads the diff and the upstream artifacts and reports…

genkovich/sdd · 81 tokens

atomic-auditor

Final gate for a finished implementation. Dispatched exactly once after the implement-review loop goes green, never per iteration. Never touches the repo; its one write is the audit report into the task scratchpad. Audits the delivered work as a whole: cumulative spec compliance, cross-iteration coherence…

damusix/atomic-claude · 169 tokens

Reviewer

Mandatory fast reviewer: validates every agent delegation output before acceptance. Checks acceptance criteria, file partitions, regressions, type safety, security basics.

monkilabs/opencastle · 30 tokens