Validate plans or specs with a multi-model council and structured verdicts.
npx skills add https://github.com/boshu2/agentops --skill pre-mortem-boshu2
Please help me install this Agent Skill. Skill: pre-mortem Source: https://github.com/boshu2/agentops/tree/main/skills/pre-mortem Command: npx skills add https://github.com/boshu2/agentops --skill pre-mortem-boshu2
Validate a plan or spec before implementation using a multi-model council to obtain early, well-reasoned judgments.
High-intent search queries and answers about installing and using this skill.
Pre-mortem plan validation is a risk assessment technique that reviews a project plan or spec before implementation. You need it to surface feasibility gaps and risks early, before committing engineers or timelines to a new feature.
To validate a plan before implementation, run a multi-model council review against your plan document. Use the /pre-mortem command with a file path to generate structured PASS/WARN/FAIL verdicts with reasoned findings.
Yes, you can run a risk assessment on a SPEC.md file. Use the /pre-mortem --deep path/to/SPEC.md command to perform a thorough feasibility review of your specification document before development begins.
The best way to review an architecture plan for risks is using /pre-mortem --preset=architecture path/to/PLAN.md. This applies a specialized council preset to surface architecture-specific feasibility gaps and produce actionable guidance.
Yes, the pre-mortem council supports quick validation. Use the /pre-mortem --quick path/to/PLAN.md command to obtain a fast, structured verdict with reasoned findings when you need rapid plan validation across workstreams.
Yes, you can trigger debate mode during plan validation by using the /pre-mortem --debate command. This engages the multi-model council in a deeper reasoning process to produce more thoroughly examined risk assessment findings.
danielcastro-devConsolidate accepted council decisions into a spec-ready draft.md.
MaxKmetGenerate structured decision briefs with verdicts, RAT experiments, pre-mortems, kill criteria, and next actions.
spideynoloveAnalyze skill directories or repositories for security risks and output a PASS/WARN/FAIL verdict.
p-sree-sai-pavanBuild minimal runnable prototypes to assess feasibility and produce a verdict.
