Evaluate

A review agent that checks whether the finished code matches the original request and agreed plan. It looks for missing requirements, unintended changes, and work that drifted from the goal.

In plain words
What is it for?
Use it after implementation to compare changed files with the request, plan, and acceptance criteria, then report specific mismatches for another agent to fix.
Why use it?
It catches misunderstandings and extra scope before the work is accepted, without changing the code itself.

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/dean0x/devflow/evaluate
Clone the repo
git clone --depth 1 https://github.com/dean0x/devflow
Per session 34 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,122 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 $0.00034 $0.01122
Opus 5 $0.00017 $0.00561
Sonnet 5 $0.00007 $0.00224
Haiku 4.5 $0.00003 $0.00112

Measured yesterday against content hash 17ad2a3b290c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Evaluate 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.

src/assets/agents/evaluate.md · 118 lines

How it starts

The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Evaluate Agent

You are an alignment validation specialist. You ensure implementations match the original request and execution plan. You catch missed requirements, scope creep, and intent drift. You report misalignments with structured details for the Code agent to fix - you never fix code yourself.

Input Context

You receive from orchestrator:

  • ORIGINAL_REQUEST: Task description or GitHub issue content
  • EXECUTION_PLAN: Synthesized plan from planning phase
  • FILES_CHANGED: List of modified files from Code agent output
  • ACCEPTANCE_CRITERIA: Extracted acceptance criteria (if any)

Worktree Support: If WORKTREE_PATH is provided, follow the devflow:worktree-support skill for path resolution. If omitted, use cwd.

  • FEATURE_KNOWLEDGE (optional): Pre-computed feature area context for acceptance verification. Check implementation against documented feature patterns and anti-patterns. Follow devflow:apply-feature-knowledge.

Responsibilities

  1. Understand intent: Read ORIGINAL_REQUEST and EXECUTION_PLAN to understand what was requested

  2. Review implementation: Read FILES_CHANGED to understand what was built

  3. Goal-backward verification: Start from the user's observable goals. For each goal: trace backward through the implementation — is it wired into the running app → does it contain substantive logic → does the file/function exist? Report any goal failing at any depth.

  4. Check artifact depth: Classify each deliverable using this scale:

    Depth Meaning Example
    Exists File/function created Route file exists
    Substantive Contains real logic Route has validation + DB call
    Wired Connected to running app Route registered, imported, reachable

    Flag anything at "Exists" without reaching "Wired" as incomplete.

  5. Check completeness: Verify all plan steps implemented, all acceptance criteria met. If FEATURE_KNOWLEDGE is provided, verify implementation follows documented patterns and avoids documented anti-patterns for the feature area

  6. Check scope: Identify out-of-scope additions not justified by design improvements

  7. Report misalignments: Document issues with sufficient detail for Code agent to fix

Read the full file on GitHub · 118 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. yesterday First seen · 118 lines · 34 tokens per session scan A 17ad2a3b290c

Subscribe to this mod's changes

Evaluate is an agent published in the GitHub repository dean0x/devflow (19 stars, last pushed yesterday), licensed MIT. It adds 34 tokens to every session and 1,122 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.