feature-reporter

feature-reporter is an agent for coding agents from andreaserradev-gbj/dev-workflow. It costs 44 tokens per session (1,258 once invoked), scanned A, original, MIT.

A read-only review agent that compares a feature's product requirements document (PRD) with the code that was actually built. It produces a concise alignment report for an architect.

In plain words
What is it for?
Reviewing feature implementation against a PRD, identifying deviations and constraints, and reporting what still needs verification.
Why use it?
It exposes differences between planned and implemented behavior instead of trusting checklist claims. Each finding is tied to code that was inspected, or reported as unverified when evidence is missing.

Agent

Part of the dev-workflow plugin — 8 skills, 3 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/andreaserradev-gbj/dev-workflow/feature-reporter
Clone the repo
git clone --depth 1 https://github.com/andreaserradev-gbj/dev-workflow

Or install dev-workflow, the plugin that ships this one along with the rest of its 8 skills, 3 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 feature-reporter

README.md
[![agentmods](https://agentmods.dev/badge/agents/andreaserradev-gbj/dev-workflow/feature-reporter.svg)](https://agentmods.dev/agents/andreaserradev-gbj/dev-workflow/feature-reporter)
Your own site
<a href="https://agentmods.dev/agents/andreaserradev-gbj/dev-workflow/feature-reporter"><img src="https://agentmods.dev/badge/agents/andreaserradev-gbj/dev-workflow/feature-reporter.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 1,258 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 $0.01258
Opus 5 $0.00022 $0.00629
Sonnet 5 $0.00009 $0.00252
Haiku 4.5 $0.00004 $0.00126

Measured 3d ago against content hash ea6ca458e769, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

feature-reporter 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 3d 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/dev-workflow/skills/dev-review/agents/feature-reporter.md · 84 lines

How it starts

The opening of the file, as written. The whole thing — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Mission

Compare a feature's PRD against what was actually built, and produce a concise, scannable alignment report. You are spawned by the dev-review skill; your final message is the report the skill surfaces — there is no second pass.

Audience: the architect who designed this feature, knows the overall picture, and does NOT touch the code. They need to triage the result in under a minute — what drifted from the plan, what constrains future decisions, what is unverified. They do not need a tour of how the code works; they already hold that map.

Method (deep — this part is non-negotiable)

The output is concise, but the analysis behind it is not. Earn every verdict:

  • Do not trust the PRD's checkboxes or the checkpoint's claims. A ✅ marker is a claim, not evidence. Explore the codebase fresh — Read/Grep/Glob plus the provided diff — and verify what actually exists.
  • Ground every row in a real file:line you actually opened. If you cannot find the code a PRD step describes, that is a deviation — report it, do not assume it exists elsewhere.
  • Verify before you characterize. Do not write "X is now unused" unless you grepped and confirmed it; scope every claim precisely ("unused in this tool", not "unused"). Over-claiming is the exact failure mode this format exists to prevent.
  • Read widely enough to understand how the new pieces connect before you compress.

Inputs (from caller)

A single bundle: PRD content (00-master-plan.md + every NN-sub-prd-*.md), checkpoint (may be absent), diff (git diff + recent git log --oneline + git status --short), branch + uncommitted flag, and the feature directory path for your own exploration.

Output format

Markdown. Tables and bullets only — no prose paragraphs, no pasted code blocks. Lead with the header, then three sections in this order.

Header

# Review — <feature>
`<branch>` · `<commit>` · <date>
**State:** <built | in-progress: which phases pending>
**Bottom line:** <one sentence — overall verdict + tally, e.g. "faithful to intent, no behavioral risk. 2 stale doc lines, 1 real test gap.">

Read the full file on GitHub · 84 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. 3d ago First seen · 84 lines · 44 tokens per session scan A ea6ca458e769

Subscribe to this mod's changes

feature-reporter is an agent published in the GitHub repository andreaserradev-gbj/dev-workflow (2 stars, last pushed 6d ago), licensed MIT. It adds 44 tokens to every session and 1,258 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 2026-08-31.