Automated summary
Invokes a deliberately contrasting advisor to critique a prior advisory answer and present the strongest opposing case on the same decision.
Strengths
- Selects a dissenting advisor based on profiles and summaries, not random choice.
- Confirms the suggested dissenter with the user and allows naming an alternative.
- Grounds the dissent in the chosen advisor's dossier and marks extrapolation.
- Refuses to manufacture conflict when the second advisor would mostly agree.
Limitations
- Requires an existing advisor library, ideally a prior /advice consult and current brief.
- The advisors are simulations in the style of real people, not the actual people.
- Relies on Claude Code-specific skills and tools such as AskUserQuestion and Skill.
Use after an advisory session when the first answer seems too clean and you want a structured, genuinely different challenge.
Avoid when you need a neutral consensus, a factual answer, or you have no advisor profiles and no context for the decision.
Security analysis
SafeThe skill uses Read/Glob/Grep/Bash only for local path setup and reading user-owned advisor/profile files, and WebSearch/WebFetch for grounding public advisor views. It contains no destructive commands, credential handling, obfuscated payloads, or instructions to disable safety. Bash appears limited to environment/path variable assignment rather than arbitrary execution.
No concerns found
How this analysis is madeExamples
/second-opinion [first-advisor] Should we prioritize enterprise sales over self-serve growth for the next two quarters?/second-opinion We're leaning toward rewriting our monolith as microservices. Get a dissenting advisor to challenge that call./second-opinion Use a skeptical operator to challenge the prior /advice recommendation to raise a large seed round.name: second-opinion description: "Get a dissenting second opinion from a DIFFERENT advisor — one likely to disagree with the first. You name the first advisor (or it uses your last /advice consult); the skill then suggests, or lets you pick, a contrasting expert and delivers their opposing view on the same decision, grounded in their public work. Use after /advice when an answer felt too clean and you want a genuinely different mind to challenge it." argument-hint: "[first-advisor] [topic] — optionally name the second advisor too" model: opus allowed-tools:
- Read
- Glob
- Grep
- Bash
- AskUserQuestion
- WebSearch
- WebFetch
- Skill
/second-opinion — A dissenting view from a different advisor
You got one advisor's take. This brings in a different advisor — deliberately chosen because they'd likely disagree — to give the opposing view on the same decision. The clash is the point: it surfaces the assumptions the first take depended on.
These are simulations in the style of real people, never the real people, and never with fabricated quotes. See $REFERENCE_DIR/voice-guide.md.
Paths
ADVISORS_HOME="${ADVISORS_HOME:-$HOME/.advisors}"
ADVISORS_DIR="$ADVISORS_HOME/advisors"
REFERENCE_DIR="$ADVISORS_HOME/reference"
STATE_DIR="$ADVISORS_HOME/state"
PROFILE="$ADVISORS_HOME/profile.md" # durable context about the person
BRIEF="$STATE_DIR/current-brief.md"
Context gate
Read $PROFILE if it exists — it tells you who you're advising (role, arena, what they optimize for); tailor the dissent to them. Then read $BRIEF if present. If the brief is missing or stale for this topic, offer to run /frame first (invoke frame via Skill). The user may skip — note the thin context if so.
How it works
1. Resolve advisor #1 and the topic
- Advisor #1 (the view to be challenged): from
$ARGUMENTS, else the advisor from the last/adviceconsult in this conversation, else the brief — otherwise ask viaAskUserQuestion. Fuzzy-match name → slug in$ADVISORS_DIR. - Topic: the decision and #1's conclusion — from
$ARGUMENTS, the conversation, or the brief'sleaning. If #1's actual conclusion isn't on record, read #1's dossier and characterize the position they'd most likely take, and say you're doing so.
2. Choose advisor #2 (the dissenter)
- Read the frontmatter (
name,role,domains) and the## Summaryof every$ADVISORS_DIR/*.md— just those (the Summary is the selection digest; fall back to frontmatter if absent). Full dossiers are loaded only for the advisor actually chosen. - Identify the advisor whose worldview most opposes advisor #1 on this specific question — different lens, different values, different track record — not just anyone.
- Suggest that advisor with a one-line reason for the expected disagreement, plus 1-2 alternative dissenters, and confirm via
AskUserQuestion(the user can pick or name their own). If the user already named a second advisor, skip the suggestion and use theirs. If the chosen advisor has no dossier, offer/advice add.
3. Deliver the dissent
- Load advisor #2's full dossier. Answer in their first-person voice.
- Open by engaging advisor #1's position directly: where #2 thinks it's wrong, and why — rooted in #2's own beliefs, frameworks, and track record.
- Make the strongest opposing case (steelman, not strawman). Name the scenario in which following advisor #1 would be a mistake, and the assumption #1's view quietly depends on.
- Stay honest: ground in the dossier, mark extrapolation, no fabricated quotes. If advisor #2 would actually largely agree with #1, say so plainly and offer a better-matched dissenter instead of manufacturing conflict.
4. Close
- The single strongest reason advisor #1 might still be right.
- The one piece of evidence that would settle the disagreement.
- Optionally suggest
/boardroomif the user wants more voices in the room.
Open with: *Second opinion — <Advisor #2>, in the style of their public work, dissenting from <Advisor #1>. Not the real people.*
Hard rules
- A different advisor, genuinely chosen to disagree — not the same person, not a random pick.
- Steelman the dissent. The opposing case must be the strongest one, argued in good faith.
- No fabricated quotes; ground each voice in its dossier; flag extrapolation.
- Don't manufacture conflict. If there's no real disagreement, say so and re-pick.
- No defamation — critiques stay real and sourced.
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/second-opinion-dissenting-view-mryiqgsa/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/second-opinion-dissenting-view-mryiqgsa' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/second-opinion-dissenting-view-mryiqgsa/download' --output '.skill-downloads/second-opinion-dissenting-view-mryiqgsa/SKILL.md.part' && mv '.skill-downloads/second-opinion-dissenting-view-mryiqgsa/SKILL.md.part' '.skill-downloads/second-opinion-dissenting-view-mryiqgsa/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.
Claude Code
test -f '.skill-downloads/second-opinion-dissenting-view-mryiqgsa/SKILL.md' && test -s '.skill-downloads/second-opinion-dissenting-view-mryiqgsa/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/second-opinion' && cp -R '.skill-downloads/second-opinion-dissenting-view-mryiqgsa' '.claude/skills/second-opinion'OpenSkills
OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.
Official OpenSkills installationopenskills install './.skill-downloads/second-opinion-dissenting-view-mryiqgsa'openskills read 'second-opinion-dissenting-view-mryiqgsa'Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.
For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.
openskills syncManual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/second-opinion-dissenting-view-mryiqgsa/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
PDF Reader
Productivity
Read and extract text from PDF files, including documents, reports, contracts, and spreadsheets. Works with local files, URLs, and WhatsApp attachments. Use whenever PDF content is needed, even if not explicitly requested.
Discord Messaging Workflows
Productivity
Use the message tool to send, edit, and moderate Discord messages with interactive components.
Interactive Verdict Preview
Productivity
Generate an interactive HTML preview of keep/drop verdicts for maintainer review, then process the pasted feedback JSON.