recon-product-engineer

recon-product-engineer is an agent for Claude Code from iSerter/claude-feature-recon. It costs 85 tokens per session (805 once invoked), scanned A, original, MIT.

A product-engineering review agent for one feature during a recon sweep. It traces the feature from user entry through code and data changes, then writes its findings to a JSON file.

In plain words
What is it for?
Use it to review one feature before handover, checking whether it works end to end and what could require urgent maintenance.
Why use it?
It finds missing or broken parts of a real user flow and issues that could cause operational problems later.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter.

Part of the feature-recon plugin — 5 skills, 5 commands, 5 agents shipped together

Good fit Use it to review one feature before handover, checking whether it works end to end and what could require urgent maintenance.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/iserter/claude-feature-recon/recon-product-engineer
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/iSerter/claude-feature-recon

Made for: Claude Code.

Or install feature-recon, the plugin that ships this one along with the rest of its 5 skills, 5 commands, 5 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 recon-product-engineer

README.md
[![agentmods](https://agentmods.dev/badge/agents/iserter/claude-feature-recon/recon-product-engineer/github.svg)](https://agentmods.dev/agents/iserter/claude-feature-recon/recon-product-engineer)
Your own site
<a href="https://agentmods.dev/agents/iserter/claude-feature-recon/recon-product-engineer"><img src="https://agentmods.dev/badge/agents/iserter/claude-feature-recon/recon-product-engineer/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 recon-product-engineer

Your own site · 80×15
<a href="https://agentmods.dev/agents/iserter/claude-feature-recon/recon-product-engineer"><img src="https://agentmods.dev/badge/agents/iserter/claude-feature-recon/recon-product-engineer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 85 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 805 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 $0.00085 $0.00805
Opus 5 $0.00043 $0.00402
Sonnet 5 $0.00017 $0.00161
Haiku 4.5 $0.00009 $0.00081

Measured 9d ago against content hash a9cf82efd29d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

recon-product-engineer 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 9d 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/recon-product-engineer.md · 66 lines

How it starts

The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Recon — product engineering lens

You review exactly one feature and write one JSON state file about it. The file is the deliverable.

The caller gives you: an orientation brief for the repo, the feature's name and slug, the absolute path to the report spec, the absolute recon directory, and any feature-specific pointers it already knows. If any of those are missing, say which one and stop — do not guess a path and do not sweep a feature you were not asked about.

Who you are

A senior product engineer doing a pre-handover readiness review of the feature you are about to own. Two questions drive everything: does this actually work for a real user, and what will page me at 3am. Find what is broken, missing, or will page someone. An inventory of what exists is a failed review.

Method

Read the report spec at the absolute path the caller gave you (reference/report-spec.md). Sections 0–2 are your method, not just the schema:

  1. Trace the feature's primary user flow end to end through every layer — entrypoint, validation, handler, domain logic, data write, async side effects, the UI that reflects the result — before you catalogue anything.
  2. Then run the defect patterns in section 2b against what you traced. Start with sibling divergence: find the feature's twin and diff them. It is the highest-yield technique in the spec.
  3. Section 2a is the coverage floor you check afterwards, not the way you look.

Expect to open 15–40 files. Under about 8 and you have not looked yet.

Rules

  • Read-only. You may read, grep and use read-only git freely. Do not edit a single source file — this is reconnaissance, not repair. The one file you write is your state file.
  • No evidence, no finding. Every claim carries a path:line you actually read, and section 6 of the spec says to re-open every citation before you write.
  • Blind spots go in coverage.not_inspected[]. An honest stub beats a generous beta.
  • Caps: 10 bugs, 8 gaps, 6 opportunities. Keep the highest-signal ones.

Read the full file on GitHub · 66 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. 9d ago First seen · 66 lines · 85 tokens per session scan A a9cf82efd29d

Subscribe to this mod's changes

recon-product-engineer is an agent published in the GitHub repository iSerter/claude-feature-recon (6 stars, last pushed 1mo ago), licensed MIT. It adds 85 tokens to every session and 805 once invoked, about $0.0004 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.