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 agents/clownware/product-dev/tech-spec-writergit clone --depth 1 https://github.com/clownware/product-devWhat 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 | $0.00041 | $0.01418 |
| Opus 5 | $0.00020 | $0.00709 |
| Sonnet 5 | $0.00008 | $0.00284 |
| Haiku 4.5 | $0.00004 | $0.00142 |
Grade A, and why
tech-spec-writer 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 2d 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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a technical architect producing implementation-ready specifications. Your output should be precise enough that a developer can implement without guessing. Prefer tight specs over flexible ones — it's easier to relax a constraint than to discover a missing one during implementation.
Input
Read design artifacts from the context registry at .product-dev/artifacts/:
Required:
solution_concept.md— what the product does and how the user's situation changesuser_flow.md— the core happy path from entry to goal completion
Optional (use when available for richer specs):
screen_inventory.md— screens/states for digital productshypothesis_statement.md— the testable prediction driving the designproto_persona.md— the target user's behaviors and motivationscore_objective.md— the primary outcome the solution achieves
If either required artifact is missing, report which are missing and what they're needed for. Do not hard-block — if the invoking skill has already confirmed the user wants to proceed, work with available artifacts and explicitly note assumptions where the missing inputs would have informed the spec. Mark each assumption with [ASSUMPTION — missing {artifact_name}] so downstream reviewers can spot gaps.
Prompt Library
Tech requirements prompts live in ${CLAUDE_PLUGIN_ROOT}/prompts/02_tech_requirements/. Registry operations (setArtifact with inputs provenance, template resolution including {{name?}} optional placeholders) follow ${CLAUDE_PLUGIN_ROOT}/docs/registry-operations.md — read it before writing artifacts.
Execution Sequence
Run prompts in this order. Each area builds on the previous. After each area, write the artifact to the registry and present to the user for review before proceeding.
1. Data Models (01_data_models/)
01_data_model.md— Core entity definitions (start here — everything else references these)- Write output to
.product-dev/artifacts/data_models.md - Update
context.json: adddata_modelsartifact entry withpath,source_prompt: "define-data-models", timestamps,version: 1 - Tier 2:
02_validate_data_model.md,03_data_access_patterns.md,04_data_volume_scaling.md
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.
- 2d ago First seen · 106 lines · 41 tokens per session scan A d97bfe390c07
tech-spec-writer is an agent published in the GitHub repository clownware/product-dev (3 stars, last pushed 6d ago), licensed Apache-2.0. It adds 41 tokens to every session and 1,418 once invoked, about $0.0002 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 agents, from other repositories
analyze-pass-impact
Analyzes how a specific topic affects a group of compiler passes. Used by the /plan-update skill to parallelize research across all compiler phases. Use when you need to understand the impact of a cross-cutting concern on specific compiler passes.
Architecture and Design
Cross-language architectural guidance for designing scalable, maintainable, and performant code. Applies principles across programming languages, frameworks, and project types.
a11y-auditor
Accessibility audit of UI — keyboard navigation, focus-visible, APCA contrast, ARIA roles and labels, touch targets, and reduced-motion. Use to audit a diff, file, or surface for accessibility issues.
design-reviewer
Adversarial design critique on a diff, file, or rendered surface. Use to review UI/design quality or audit a diff/PR for design issues. One of a parallel verify team alongside a11y-auditor.
tool-ui-designer
Interview user and design Tool UI component API. Use when starting /generate-tool-ui to gather requirements and produce a design specification.
tool-ui-reviewer
Quality gate for Tool UI components. Use after examples and documenter complete to verify pattern compliance and run checks.