Reality Check

Prevent bad strategic decisions by surfacing assumptions, testing feasibility, and exposing critical flaws before execution.

Security Scan

The skill is an instruction-only feasibility/assumption-checker whose declared purpose, inputs, and runtime instructions are internally consistent and do not request extra access or install anything.

Purpose & Capability

Name, description, and runtime instructions all describe the same capability (surfacing assumptions, proposing tests, scoring viability). The skill requires no binaries, credentials, or config paths — consistent with a lightweight analytical helper.

Instruction Scope

SKILL.md steps are limited to identifying assumptions, proposing tests, assessing impact, and producing a score/recommendation. The instructions do not direct the agent to read files, access external endpoints, or harvest environment data. The only open behavior is 'extract assumptions' when not provided, which is reasonable for this task.

Install Mechanism

There is no install spec and no code files — the skill is instruction-only, so nothing is written to disk or fetched at install time.

Credentials

The skill declares no required environment variables, credentials, or config paths. There is no disproportionate request for secrets or unrelated service access.

Persistence & Privilege

The skill is not marked always:true and uses normal autonomous invocation defaults. It does not request persistent system-level privileges or attempt to modify other skills' configurations.

Assessment

This skill appears coherent and low-risk because it is instruction-only and asks for no external credentials or installs. Before relying on its outputs: (1) treat viability scores and recommendations as advisory, not definitive—verify critical conclusions with humans, especially for legal/financial decisions; (2) test the skill on low-risk/example ideas to confirm the style and thoroughness of the extracted assumptions/tests; (3) avoid submitting sensitive secrets or private data into the prompts; and (4) note the source is 'Morpheus' / unknown registry metadata—if provenance matters for your use case, verify author/trust independently.

SKILL: reality-check

Purpose

Prevent bad strategic decisions by forcing assumptions into the open, testing them, and ranking feasibility before commitment.

When to Use

• A plan sounds good but may be fragile or unrealistic
• The user is optimistic without evidence
• A decision has meaningful downside risk

Inputs

• idea (required): the proposed plan/decision
• assumptions (optional): stated assumptions; if missing, Morpheus must extract them

Steps

1. List assumptions (explicit + implicit).
2. For each assumption:
   • define what would make it true/false
   • define the cheapest test/experiment
   • define failure impact if wrong
3. Identify the weakest links (highest impact × lowest evidence).
4. Evaluate feasibility:
   • resources/time
   • constraints/governance
   • reversibility
5. Produce:
   • viability score (0–100)
   • critical flaws
   • recommendation (proceed / revise / stop)
6. Provide the next 1–3 validation steps.

Validation

• Assumptions are explicit and testable.
• Claims are labeled as evidence vs hypothesis.
• Recommendation follows the evidence, not optimism.

Output

• viability_score (0–100)
• critical_flaws
• recommendation
• next_steps

Safety Rules

• Prioritize truth over optimism.
• No financial guarantees or claims of certainty without evidence.

Example

Idea: “Launch a new token feature in 48h.” Output: viability 35/100; critical flaw: no governance approval + no test coverage; next steps: scope reduction + emergency review + paper validation.