From structured-english
Generates a human-readable SESF v4.1 specification for a given domain or topic using formal BEHAVIOR, PROCEDURE, RULE, and STEP blocks with rationale annotations.
How this command is triggered — by the user, by Claude, or both
Slash command
/structured-english:write-human-spec <domain or topic to specify>Files this command reads when invoked
This command is limited to the following tools:
The summary Claude sees in its command listing — used to decide when to auto-load this command
# Write a Human-Facing Specification Generate a complete SESF v4.1 specification optimized for human readers and maintainers. The output uses formal BEHAVIOR/PROCEDURE/RULE/STEP blocks that provide visual scaffolding humans can scan, review, and modify. Includes rationale annotations explaining *why* rules exist. ## Audience The primary reader of this spec is a human — an engineer, product manager, or domain expert who needs to understand, review, or modify the spec. Optimize for: - **Comprehension** — formal blocks categorize concerns visually; rationale annotations explain the *why* -...
Generate a complete SESF v4.1 specification optimized for human readers and maintainers. The output uses formal BEHAVIOR/PROCEDURE/RULE/STEP blocks that provide visual scaffolding humans can scan, review, and modify. Includes rationale annotations explaining why rules exist.
The primary reader of this spec is a human — an engineer, product manager, or domain expert who needs to understand, review, or modify the spec. Optimize for:
Use the sesf skill — it contains all the rules, formats, and validation requirements for SESF v4.1. Read it fully before proceeding. Also read:
${CLAUDE_PLUGIN_ROOT}/skills/sesf/assets/reference.md${CLAUDE_PLUGIN_ROOT}/skills/sesf/assets/authoring-guide.md — this walks through the thinking processWalk the user through the thinking process from the authoring guide:
**BEHAVIOR** blocks)**PROCEDURE** blocks)Based on the requirements, select the appropriate tier:
Present the plan to the user before writing:
${CLAUDE_PLUGIN_ROOT}/skills/sesf/assets/template.md${CLAUDE_PLUGIN_ROOT}/skills/sesf/references/examples.md for the selected tier**BEHAVIOR** name: description blocks**RULE** name: with WHEN condition THEN action syntax inside behaviors**PROCEDURE** name: description blocks**STEP** name: description with optional → $variable for data flow## Errors table at the endPresent the finished specification to the user for review. Highlight design decisions (e.g., "I separated validation into its own BEHAVIOR block because those rules apply across multiple procedures").
Run the structural validator:
python3 ${CLAUDE_PLUGIN_ROOT}/skills/sesf/scripts/validate_sesf.py <output-file>
Show the validation results to the user. If there are failures, present each proposed fix and ask the user to approve before applying.
Ask the user where to save the specification, or suggest a default based on the domain name (e.g., <domain>-spec.md). If the target file already exists, ask the user whether to overwrite it or save to a different path.
npx claudepluginhub reggiechan74/cc-plugins --plugin structured-english/specAuthors structured NLSpec from natural language description using multi-AI research, with adversarial completeness validation, scoring, and saved spec file.
/specsGenerates collaborative specifications for a feature, producing Intent, BDD scenarios, architecture notes, and acceptance criteria.
/specGenerates spec-driven development docs from a feature description: requirements.md (EARS user stories), design.md (Mermaid diagrams, TypeScript interfaces), tasks.md (implementation plan).
/specGenerates spec docs from feature description: requirements.md (EARS notation), design.md (Mermaid diagrams, TS interfaces), tasks.md (implementation plan).
/spec-delegatedGenerates a delegated execution spec from the current brainstorming session and saves it as specs/YYYY-MM-DD-<name>.md using today's date.