Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/jimmc414/claude-code-plugin-marketplace/collaborative-spec-buildergit clone --depth 1 https://github.com/jimmc414/claude-code-plugin-marketplaceWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/jimmc414/claude-code-plugin-marketplace/collaborative-spec-builder)<a href="https://agentmods.dev/commands/jimmc414/claude-code-plugin-marketplace/collaborative-spec-builder"><img src="https://agentmods.dev/badge/commands/jimmc414/claude-code-plugin-marketplace/collaborative-spec-builder.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00012 | $0.00848 |
| Opus 5 | $0.00006 | $0.00424 |
| Sonnet 5 | $0.00002 | $0.00170 |
| Haiku 4.5 | $0.00001 | $0.00085 |
Grade A, and why
collaborative-spec-builder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Collaborative Specification Builder
You are entering a collaborative specification building session with the user. This is an iterative process to define an unambiguous specification before implementation.
Specification Areas to Cover
Systematically gather information across these areas:
Intent
- What problem does this solve? Why does it need to exist?
- What is the broader context? What goals guide decisions?
- What is explicitly out of scope? How will this be used?
Domain Types / Vocabulary
- What are the core types involved?
- What do key terms mean precisely?
Input Contract
- What inputs are accepted? Types, formats, valid ranges?
Output Contract
- What is returned? Shape/type of output?
Preconditions
- What must be true before execution?
- What is validated vs. assumed?
Postconditions
- What is guaranteed after successful execution?
Invariants
- What must remain true throughout execution?
Constraints
- What rules govern valid behavior? What is prohibited?
State Transitions
- What states can entities be in? What transitions are legal?
Side Effects
- What external state is modified? What I/O occurs?
Error Cases
- What can fail? How should each failure be handled?
Boundary Conditions
- Empty inputs, nulls, zeros, max values?
Assumptions
- What is assumed true but not explicitly checked?
Dependencies / Environment (if relevant)
- Required libraries, services, runtime, configuration?
Non-functional Requirements (if relevant)
- Performance, memory, scalability expectations?
Security (if relevant)
- Auth requirements, data sensitivity, threat model?
Concurrency (if relevant)
- Parallel execution, synchronization, race conditions?
Examples / Acceptance Criteria
- Concrete input/output pairs? Critical test cases?
Phase 1: Specification Gathering (Iterative Q&A)
Begin by understanding what the user wants to build through iterative questioning:
- Use the AskUserQuestion tool to ask clarifying questions
- You may ask up to 4 questions per round (this is the tool's limit)
- Focus each round on specific specification areas that are unclear or undefined
- IMPORTANT: Always include this option in every question round:
- Label: "Done - specification is complete"
- Description: "The specification is sufficiently clear, proceed to planning"
- After receiving answers, analyze the responses and update your understanding
- If the user did NOT select "Done", continue asking follow-up questions
- Repeat until the user selects "Done - specification is complete"
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 119 lines · 12 tokens per session scan A 6e0fa1466682
collaborative-spec-builder is a command published in the GitHub repository jimmc414/claude-code-plugin-marketplace (4 stars, last pushed today), licensed MIT. It adds 12 tokens to every session and 848 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.