second-opinion

second-opinion is a skill for Claude Code from SinanTufekci/agent-intern. It costs 86 tokens per session (853 once invoked), scanned A, original, MIT.

A skill that asks a separate coding agent for an independent review of a change, plan, design, or bug, then checks its findings against the code. It can use agents connected through agent-intern.

In plain words
What is it for?
Use it before merging a risky change, when debugging is difficult, or when someone requests a second review. It requests concrete failure scenarios and confidence levels for findings.
Why use it?
It reduces the risk of missing a bug or accepting one model's opinion without checking the evidence.

Skill for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: mentions Codex.

Part of the agent-intern plugin — 3 skills, 1 MCP server shipped together

Good fit Use it before merging a risky change, when debugging is difficult, or when someone requests a second review. It requests concrete failure scenarios and confidence levels for findings.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/sinantufekci/agent-intern/second-opinion
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.

Any agent
npx skills add SinanTufekci/agent-intern --skill second-opinion
Clone the repo
git clone --depth 1 https://github.com/SinanTufekci/agent-intern

Made for: Claude Code.

Or install agent-intern, the plugin that ships this one along with the rest of its 3 skills, 1 MCP server.

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 second-opinion

README.md
[![agentmods](https://agentmods.dev/badge/skills/sinantufekci/agent-intern/second-opinion/github.svg)](https://agentmods.dev/skills/sinantufekci/agent-intern/second-opinion)
Your own site
<a href="https://agentmods.dev/skills/sinantufekci/agent-intern/second-opinion"><img src="https://agentmods.dev/badge/skills/sinantufekci/agent-intern/second-opinion/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 second-opinion

Your own site · 80×15
<a href="https://agentmods.dev/skills/sinantufekci/agent-intern/second-opinion"><img src="https://agentmods.dev/badge/skills/sinantufekci/agent-intern/second-opinion.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 86 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 853 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 $0.00086 $0.00853
Opus 5.5 $0.00034 $0.00341
Sonnet 5.5 $0.00017 $0.00171
Haiku 4.5 $0.00009 $0.00085

Measured 4d ago against content hash a77cf6088ae1, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-29, from the pricing page.

Security

Grade A, and why

second-opinion 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 4d 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.

plugin/skills/second-opinion/SKILL.md · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Get a second opinion on: $ARGUMENTS

If that is empty, review the working tree's uncommitted changes. If you invoked this skill on your own and the user didn't ask for a review, ask them in one line before spending their quota.

1. Build the brief

Put everything the reviewer needs inline in the prompt. Don't rely on it reading files: some sandboxes can't. Codex on Windows is the known case, because its read-only sandbox currently refuses every command.

  • The change. Use git diff HEAD, or the plan, design or bug the user pointed at. If the diff is over ~1,500 lines, keep the most relevant files and say which ones you left out.
  • One paragraph of context. What the change is meant to do, and anything the reviewer can't infer from the diff.
  • Not your opinion. Leave out what you think of the change. The value is an independent read.

Ask the reviewer for correctness bugs first, each with a file:line and a concrete failing scenario. Then risky edge cases, then anything it would do differently. It should rate each finding high, medium or low confidence, and say "no issues found" rather than invent some.

2. Pick reviewers

  • Default: one reviewer from a different model family than you. Try them in this order:
    • codex_ask
    • copilot_ask or cursor_ask
    • antigravity_ask
    • opencode_ask. Its free models are slow, so warn the user it can take several minutes.
  • --council, or the user wants several opinions: run the council preset: preset_swarm(preset="council", material=<the brief from step 1>, workspace=<repo root>). It runs Codex, Copilot and Gemini reviewers read-only, in parallel, each told to work independently. A reviewer whose backend isn't set up fails in place and the others still answer. If fewer than two answer, run an agent_swarm with the reviewers that do work. The user may have their own council preset with other reviewers; swarm_presets shows which one will run.
  • Keep it read-only. Use sandbox: "read-only" wherever a backend accepts it, and set workspace to the repo root. For Antigravity, pass plan: true to antigravity_ask, or sandbox: "read-only" in a swarm task, so it can't edit files.
  • If a backend isn't set up, the failed call says so. Fall back to the next reviewer instead of stopping. If none work, suggest /agent-intern:doctor.

Read the full file on GitHub · 61 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. 4d ago First seen · 61 lines · 86 tokens per session scan A a77cf6088ae1

Subscribe to this mod's changes

second-opinion is a skill published in the GitHub repository SinanTufekci/agent-intern (25 stars, last pushed 5d ago), licensed MIT. It adds 86 tokens to every session and 853 once invoked, about $0.0003 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 2026-09-25.

Related

Other skills, from other repositories

doubt-driven-development

Subjects every non-trivial decision to a fresh-context adversarial review before it stands. Use when you want every assumption cross-examined before proceeding, when stress-testing a plan for hidden failure modes, when correctness matters more than speed, when working in unfamiliar code, when stakes are high…

addyosmani/agent-skills · 96 tokens

conductor-review

Reviews the completed track work against guidelines and the plan. Acts as a Principal Software Engineer to ensure quality and compliance.

gemini-cli-extensions/conductor · 27 tokens

graph-it-live

Analyze code dependencies, call graphs, and architecture using the graph-it CLI. Use this skill whenever the user wants to understand how their code is connected — even if they don't say "dependency graph" explicitly. Trigger for: "what calls this function", "who uses this class", "is it safe to delete X", "what…

magic5644/Graph-It-Live · 165 tokens

debugger

Rank root-cause hypotheses and propose the smallest safe fix.

antonbabenko/deliberation · 15 tokens

stage-refactor-checklist

Comprehensive checklist for refactoring a schemacompiler stage folder. Use EVERY TIME you refactor a compiler stage (resolve, unify, codegen, etc.). Read this before writing any code.

Kevin-Liu-01/Agent-Machines · 45 tokens

invariant-first-coding

Reason from proven invariants before coding. Use when reviewing fallback logic, narrowing types, handling conversions, deciding whether an error path is real or impossible, removing overdefensive code, or when the user questions whether code is being too theatrical, too hedged, or not fail-closed enough.

Kevin-Liu-01/Agent-Machines · 65 tokens