Plan red-team review
Assign a plan, PRD, or strategy and a red-team agent grills its load-bearing assumptions, runs a pre-mortem, and hands you a ranked risk memo with the cheapest test for each weakness before you commit to building.
Saves you ~3.7 h / run
How it works
- Trigger
- When you start the “Red-team plan” workflow.
- Job
- Grill to understanding, attack the load-bearing assumptions, run a pre-mortem.
- Outcome
- A ranked risk memo with mitigations and decision points.
What it installs
Agents 2
-
Red-Team Agent
Grills plans, runs pre-mortems, and produces ranked, actionable risk memos.
-
Plan red-team review Quality Reviewer
Checks primary evidence, domain controls, deliverable completeness, and communication quality, stopping the run when the work is wrong, unsupported, incomplete, or uncertain.
Teams 1
-
Plan red-team review quality team
The delivery agents produce the work while an independent quality reviewer checks each workflow handoff against explicit evidence, domain, and communication requirements before the run can continue.
Workflows 1
-
Red-team plan
Grill to understanding, attack the load-bearing assumptions, run a pre-mortem, self-critique, then hand the ranked risk memo off for approval.
Skills 3
-
grilling
Interview a plan or design relentlessly to reach a shared understanding before building — one question at a time, walking each branch of the design tree, resolving dependencies in order, and reading the plan or codebase instead of asking when the answer is already there. Adapted from mattpocock/skills/grilling.
-
strategy-red-team
Red-team a plan, PRD, or strategy by attacking its load-bearing assumptions before reality does — steelman each claim, then attack it, write each failure as 'Fails if ___', rank by impact × likelihood × cheapness-to-test, and return the cheapest test and kill criterion for each. Adapted from phuryn/pm-skills/strategy-red-team.
-
pre-mortem
Run a pre-mortem on a plan or launch by imagining it already failed and working backward — categorize risks as Tigers (real), Paper Tigers (overblown), and Elephants (unspoken), classify Tigers as launch-blocking, fast-follow, or track, and give each launch-blocker a mitigation, owner, and date. Adapted from phuryn/pm-skills/pre-mortem.
Setup guide
How to Red-Team a Plan With an Agent
A practical guide to pressure-testing plans, PRDs, and strategies with assumption grilling, pre-mortems, and approval.
Read the setup guideDon't see your workflow? Describe it.
A sentence or two about a recurring job is enough. We design the playbook that runs it and show you exactly what it saves.