reflect

reflect is a command for coding agents from JoelYYoung/retrolens. It costs 0 tokens per session (538 once invoked), scanned A, original, MIT.

A command that reviews an analyzed AI-agent session and turns its useful patterns into notes. It produces lessons for people and instructions for future agents.

In plain words
What is it for?
Use it after session analysis to write lessons to a Markdown or JSON file, either for a chosen session or the latest one.
Why use it?
It saves you from manually searching a long conversation for errors, wasted steps, successful practices, and environment problems.

Command

Part of the retrolens plugin — 1 skill, 4 commands 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 commands/joelyyoung/retrolens/reflect
Clone the repo
git clone --depth 1 https://github.com/JoelYYoung/retrolens

Or install retrolens, the plugin that ships this one along with the rest of its 1 skill, 4 commands.

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 reflect

README.md
[![agentmods](https://agentmods.dev/badge/commands/joelyyoung/retrolens/reflect.svg)](https://agentmods.dev/commands/joelyyoung/retrolens/reflect)
Your own site
<a href="https://agentmods.dev/commands/joelyyoung/retrolens/reflect"><img src="https://agentmods.dev/badge/commands/joelyyoung/retrolens/reflect.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 538 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 5.1 $0.00000 $0.00538
Opus 5 $0.00000 $0.00269
Sonnet 5 $0.00000 $0.00108
Haiku 4.5 $0.00000 $0.00054

Measured 5d ago against content hash 9ab9ca4efe60, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

reflect 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 5d 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.

retrolens-plugin/commands/reflect.md · 68 lines

How it starts

The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.

retrolens:reflect Command

Extract reusable human lessons and agent directives from a session analysis.

Full skill reference: Read ./SKILL.md → Workflow D for lesson formats and writing standards.

Prerequisites

Run /retrolens:analyze first to produce a session analysis.

Usage

/retrolens:reflect [<session-id>] [--out <file>] [--format <md|json>]

Arguments

  • <session-id> - Session ID/prefix, latest, or omit to use the most recently analyzed session

Options

  • --out <file> - Write output to a file (default: notes/session-<ID>-lessons.md)
  • --format <md|json> - Output format (default: md)

What This Command Does

Step 1: Read the session through learning lenses

For each turn, identify:

Signal What to look for
🔴 Errors & Fixes Tool failures, repeated retries, user corrections
🟡 Inefficiency Unnecessary exploration, redundant reads, off-target tool calls
🟢 Effective practices Right tool first try, clean drill-down, good commit points
⚠️ Environment traps API limits, version issues, proxy failures

Step 2: Produce two output sections

Human lessons — how to interact with agents better:

  • Which prompts worked first try vs. caused confusion?
  • What upfront context would have saved turns?
  • Which tasks were too large and should be split?

Agent directives — reusable rules for future sessions:

  • Project conventions (naming, file structure, import order)
  • Environment gotchas (proxy, API keys, version constraints)
  • Sequencing rules ("always run tests after editing")
  • Known pitfalls discovered in this session

Step 3: Write the artifact

Ask the user where to save:

Where should I write the lessons?

  1. notes/session-<ID>-lessons.md — standalone file
  2. Append to AGENTS.md — VS Code Copilot instructions
  3. Append to CLAUDE.md — Claude Code instructions
  4. Append to .github/copilot-instructions.md — Copilot global
  5. Custom path

Read the full file on GitHub · 68 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. 5d ago First seen · 68 lines · 0 tokens per session scan A 9ab9ca4efe60

Subscribe to this mod's changes

reflect is a command published in the GitHub repository JoelYYoung/retrolens (2 stars, last pushed 4mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 538 tokens. 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.