workshop / day-3

Day 3 · Wednesday, 3 June 2026

Customize & extend

Deliverable
Two-subagent critique of your Day 2 regression (one a causal-inference critic) + a three-subagent research report + 2 MCP servers + Superpowers + plugin-dev + Crawfurd's paper-review skill installed + a Superpowers-driven spec for an extended CPD analysis + your own plugin generated via /plugin-dev:create-plugin.

Slides

Open deck in presentation mode ↗

Agenda

  1. § 1
    Day 2 recap + Day 3 framing
    Discussion
  2. § 2
    Subagents + hostile reviewer
    Lecture
  3. § 3
    Dispatch a subagent to critique your §4 regression

    Launch two subagents to critique your empirical work

    1. 1
      Have the agent dispatch two fresh-context subagents and point both at your empirical work.
    2. 2
      Brief one as a generic hostile reviewer.
    3. 3
      Brief the other with a sharper lens of your choice: a causal-inference critic, a mechanism critic, whatever fits your work.
    4. 4
      Have the main agent summarize the critiques. You decide what to act on.

    Launch three subagents to research a topic, then have the main agent write a report

    1. 1
      Pick a topic you want to understand.
    2. 2
      Have the main agent dispatch three subagents, each on a distinct angle.
    3. 3
      Have it synthesize their findings into one short report.
    Done early? Try one of these.
    1. 1
      Chain the hostile reviewer into a fixer: feed its critique to a second subagent that proposes concrete fixes.
    2. 2
      Pit a critic against a defender, then adjudicate the disagreement yourself.
    3. 3
      Run a hostile and a sympathetic reviewer on the same work and compare what each surfaces.
    4. 4
      Red-team your own project memory file: have a subagent attack your conventions for gaps and contradictions.
    Workshop
  4. § 4
    MCP servers
    Lecture
  5. § 5
    Install + use Playwright MCP and a docs-search MCP

    Talk to the agent about

    1. 1
      Install the Playwright MCP server (github.com/microsoft/playwright-mcp).
    2. 2
      Install a docs-search MCP such as Context7 (github.com/upstash/context7-mcp).
    3. 3
      Verify both show as connected.
    4. 4
      Verify Playwright works: ask the agent to open a real web page and report what's there.
    5. 5
      Verify Context7 works: ask for the current signature of a fixest or pandas function and confirm it cites the doc source.
    Workshop
  6. § 6
    Plugins + the Superpowers ecosystem
    Lecture
  7. § 7
    Install Superpowers + plugin-dev + Crawfurd's paper-review skill

    Talk to the agent about

    1. 1
      Add the official plugin marketplace if it isn't already configured.
    2. 2
      Install the plugin builder your agent uses. You author your own plugin with it later.
    3. 3
      Install Superpowers (github.com/obra/superpowers).
    4. 4
      Install Crawfurd's paper-review skill (lcrawfurd.github.io/claude-skills).
    5. 5
      Verify each install: ask the agent to list available plugins and skills.
    Workshop
  8. § 8
    Spec-driven development
    Lecture
  9. § 9
    Use Superpowers to spec the CPD analysis

    Talk to the agent about

    1. 1
      Invoke Superpowers' brainstorming skill to spec an extended CPD analysis that builds on your Day 2 regression.
    2. 2
      Let it interview you, one question at a time.
    3. 3
      When it asks if you're ready, switch to writing-plans to turn the spec into an analysis plan.
    4. 4
      Save the plan and read it end to end before approving.
    Workshop
  10. § 10
    Plugin builder + lor preview
    Lecture
  11. § 11
    Create your own plugin with /plugin-dev:create-plugin

    Talk to the agent about

    1. 1
      Use the plugin-dev plugin to create your own plugin: run /plugin-dev:create-plugin and let it interview you about what the plugin should do.
    Stuck for an idea? Build one of these.
    1. 1
      Revise-and-resubmit responder: /rr:start runs one subagent per reviewer to role-play their objections, drafts the point-by-point reply in your voice, and an adversarial subagent checks each reply closes the point. A hook protects the submitted manuscript.
    2. 2
      Replication-package builder: /repkit:build orders your scripts into a master run, one subagent audits for nonreproducible code while another regenerates and diffs every table and figure, and a hook blocks done until they match.
    3. 3
      Pre-analysis plan: /pap:draft interviews you on hypotheses and design, subagents enumerate the specification grid and multiple-testing corrections, a power skill simulates minimum detectable effects, and it compiles a timestamped OSF-ready PAP.
    4. 4
      Discussant report: /discuss:prep reads a paper you are assigned to discuss, subagents extract the contribution and stress-test the identification, and a skill drafts your discussant slides and speaking notes.
    5. 5
      Grant proposal builder: /grant:draft drafts specific aims, broader impacts, and the data-management plan with phased subagents, and a critic subagent scores it against the solicitation's review criteria.
    6. 6
      Letter-of-recommendation writer: /lor:write interviews you about the candidate, drafts in your voice from past letters, a subagent calibrates the signal to the strength you intend, and it compiles to a letterhead PDF.
    Workshop
  12. § 12
    Odds and Ends
    Lecture
  13. § 13
    Debrief + Day 4 preview
    Discussion

Materials

← All days