Requirement to tlaplus property generator
Skill ArabelaTso/Skills-4-SE/skills/requirement-to-tlaplus-property-generator
A curated list of 180+ useful Claude Skills for Software Engineering and resources for customizing AI for SE workflows.
npx -y skills add ArabelaTso/Skills-4-SE --skill requirement-to-tlaplus-property-generatorAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
What its author says it does
Copied from the file, not written here
Automatically derives TLA+ properties (invariants, safety, liveness) from natural-language requirements or structured requirement documents. Resolves ambiguities, asks clarifying questions for underspecified requirements, and outputs TLA+-compatible property definitions with semantic explanations. Use when translating system requirements, specifications, or behavioral constraints into formal TLA+ temporal logic properties for verification with TLC model checker.
SKILL.md
7.4 KB, as published. Nobody here has run it
Requirement to TLA+ Property Generator
Overview
This skill transforms natural-language requirements into formal TLA+ properties, enabling rigorous verification of system specifications. It handles invariants, safety properties, and liveness properties, resolving ambiguities and seeking clarification when needed.
Workflow
Step 1: Analyze the Requirement
Parse and understand the input requirement:
-
Identify requirement type:
- Invariant: "The system always maintains property P"
- Safety: "Bad thing X never happens"
- Liveness: "Good thing Y eventually happens"
- Fairness: "Process Z gets a fair chance"
-
Extract key elements:
- State variables involved
- Conditions and constraints
- Temporal aspects (always, eventually, until, etc.)
- Quantifiers (for all, exists)
-
Detect ambiguities:
- Undefined terms or variables
- Unclear temporal scope
- Missing boundary conditions
- Implicit assumptions
Step 2: Resolve Ambiguities
When ambiguities are detected, use one of these strategies:
Strategy A: Reasonable Interpretation If the ambiguity has a standard interpretation in the domain, resolve it automatically and document the assumption.
Example:
- Requirement: "The buffer never overflows"
- Interpretation: "buffer_size <= MAX_BUFFER_SIZE" (assuming MAX_BUFFER_SIZE is defined)
Strategy B: Ask Clarifying Questions For critical or non-obvious ambiguities, ask the user:
- "Does 'eventually' mean within a bounded time or unbounded?"
- "Should this property hold for all processes or just specific ones?"
- "What is the initial state assumption?"
See references/clarification_patterns.md for common question templates.
Step 3: Map to TLA+ Constructs
Translate the requirement into TLA+ syntax:
Invariants (Type Invariant or State Predicate):
TypeOK == /\ var1 \in ValidSet1
/\ var2 \in ValidSet2
/\ condition
StateInvariant == predicate_over_state_variables
Safety Properties (Always):
Safety == []P \* P holds in all reachable states
Liveness Properties (Eventually):
Liveness == <>P \* P eventually holds
Liveness2 == [](P => <>Q) \* If P holds, Q eventually holds
Fairness:
Fairness == WF_vars(Action) \* Weak fairness
Fairness2 == SF_vars(Action) \* Strong fairness
See references/tlaplus_syntax.md for comprehensive syntax reference.
Step 4: Generate TLA+ Property Definition
Create the complete property definition:
- Choose appropriate name: Descriptive and follows TLA+ conventions
- Write the formula: Use correct TLA+ temporal operators
- Add comments: Explain the property's intent
- Include in spec: Show where to place it in the TLA+ module
Example output:
\* Property: Buffer never overflows
\* Requirement: "The system must ensure the buffer size never exceeds capacity"
\* Type: Safety (Invariant)
BufferSafety == buffer_size <= MAX_BUFFER_SIZE
\* Add to specification:
\* Spec == Init /\ [][Next]_vars /\ BufferSafety
Step 5: Provide Semantic Explanation
Explain what the property means:
- Plain English: Restate in clear, unambiguous language
- Violation scenario: Describe what would violate this property
- Verification approach: How TLC will check it
- Related properties: Mention dependencies or related properties
Common Requirement Patterns
Pattern 1: Mutual Exclusion
Requirement: "At most one process can be in the critical section"
TLA+ Property:
MutualExclusion ==
\A p1, p2 \in Processes :
(p1 # p2) => ~(pc[p1] = "critical" /\ pc[p2] = "critical")
Pattern 2: Eventual Progress
Requirement: "Every request is eventually served"
TLA+ Property:
EventualService ==
\A req \in Requests :
(req.status = "pending") ~> (req.status = "completed")
Pattern 3: Bounded Response
Requirement: "The system responds within N steps"
TLA+ Property:
\* Note: TLA+ doesn't directly express bounded liveness
\* Use auxiliary counter variable
BoundedResponse ==
[](request_made => <>(response_sent \/ timeout_counter > N))
Pattern 4: Ordering Constraint
Requirement: "Event A must occur before event B"
TLA+ Property:
OrderingConstraint ==
[](event_B_occurred => event_A_occurred_before)
See references/requirement_patterns.md for more patterns.
Handling Complex Requirements
Compound Requirements
Break down complex requirements into multiple properties:
Requirement: "The system must process requests in FIFO order and complete each within 10 steps"
Decomposition:
- FIFO ordering property
- Bounded completion property
- Conjunction of both
Conditional Requirements
Use implication for conditional properties:
Requirement: "If the system is in safe mode, no writes are allowed"
TLA+ Property:
SafeModeConstraint == [](safe_mode => ~write_enabled)
Quantified Requirements
Use TLA+ quantifiers appropriately:
Requirement: "All active processes eventually terminate"
TLA+ Property:
AllTerminate ==
\A p \in Processes :
(status[p] = "active") ~> (status[p] = "terminated")
Best Practices
- Start simple: Begin with basic invariants before complex temporal properties
- Be explicit: Document all assumptions and interpretations
- Use meaningful names: Property names should reflect their purpose
- Test incrementally: Verify simple properties first
- Consider negation: Sometimes it's easier to express "bad thing never happens"
- Check consistency: Ensure properties don't contradict each other
Common Pitfalls
Pitfall 1: Confusing safety and liveness
- Safety: "Nothing bad happens" ([]P)
- Liveness: "Something good happens" (<>P)
Pitfall 2: Unbounded liveness without fairness
- Liveness properties often require fairness assumptions
Pitfall 3: Over-specification
- Too many constraints may make the spec unrealizable
Pitfall 4: Implicit state assumptions
- Always make initial state assumptions explicit
Output Format
For each requirement, provide:
- Original Requirement: Quote the input
- Property Type: Invariant/Safety/Liveness/Fairness
- TLA+ Definition: Complete, syntactically correct property
- Semantic Explanation: What it means in plain language
- Assumptions: Any assumptions made during translation
- Integration: How to add it to the TLA+ specification
- Verification Notes: Tips for checking with TLC
References
- references/tlaplus_syntax.md: Complete TLA+ temporal logic syntax
- references/requirement_patterns.md: Common requirement-to-property mappings
- references/clarification_patterns.md: Question templates for ambiguous requirements