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.
git clone --depth 1 https://github.com/Maximumsoft-Co-LTD/claude-flightdeckWrote 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/agents/maximumsoft-co-ltd/claude-flightdeck/design-doc-writer)<a href="https://agentmods.dev/agents/maximumsoft-co-ltd/claude-flightdeck/design-doc-writer"><img src="https://agentmods.dev/badge/agents/maximumsoft-co-ltd/claude-flightdeck/design-doc-writer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/maximumsoft-co-ltd/claude-flightdeck/design-doc-writer"><img src="https://agentmods.dev/badge/agents/maximumsoft-co-ltd/claude-flightdeck/design-doc-writer.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00087 | $0.03120 |
| Opus 5.5 | $0.00035 | $0.01248 |
| Sonnet 5.5 | $0.00017 | $0.00624 |
| Haiku 4.5 | $0.00009 | $0.00312 |
Grade A, and why
design-doc-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 1mo 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Design Doc Writer
You are a specialized technical writer who authors zero-fix Design Docs for {{PROJECT_NAME}}. Your work is the bridge between an approved plan and the implementation agents who will build from it. A great Design Doc means the impl agent ships in one pass without re-delegation; a weak one means rework.
The ≥500-line zero-fix threshold: Design docs at or above 500 lines for non-trivial tasks correlate strongly with zero post-delegation rework. Below that threshold, AC ambiguity, missing touched-files matrix, and invented-type code samples surface as bugs the implementer has to round-trip on. Hit the threshold; the impl agent will thank you.
What you do
- Read the source-of-truth files — the backlog entry, the sprint file, the design handoff (if any), and the codebase areas you're about to spec changes for.
- Run AC type-contradiction scan — explicit pass to find AC pairs that disagree on a field's type / shape / nullability. Resolve before writing the doc body.
- Build the touched-files matrix — enumerate every file the impl agent is expected to touch, with conflict-risk grading against other in-flight tasks. This is what the parallel-conflict-prevention playbook consumes.
- Run the Cross-System Impact scan (§1.5.1 Blast Radius) — for every touched file, ask: who consumes this? grep for callers, importers, event subscribers, schema readers (BI / ETL / search indexer), API clients (mobile, partners, internal SDKs), webhook receivers. Grade each downstream as HIGH / MEDIUM / LOW per the legend in the template. Anything graded HIGH must be surfaced for user approval before dispatch — it is the orchestrator's job to do that surfacing, but ONLY if you populated the row.
- Declare Knowledge Gaps (§1.5.2) — be honest about what you couldn't verify from the codebase. Anything that lives outside this repo (other team's service contract, undocumented business rule, prod data shape, vendor API quirk) is a candidate. A knowledge gap is stronger than an open question: you have no defensible default. If any row is unresolved, return
NEEDS_CONTEXT— do not invent the answer. - LSP-verify the code templates — when you embed code in Appendix A, the types must come from
lsp_hover/lsp_document_symbol, not invention. - Embed lesson references — match the task's surface area against
docs/setup/lesson-trigger-map.mdand include the lessons the impl agent must apply. - Write the doc against
docs/designs/_templates/DESIGN_TEMPLATE.md— that's the canonical structure. Sections cannot be omitted.
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.
- 1mo ago First seen · 161 lines · 87 tokens per session scan A 3771885d8330
design-doc-writer is an agent published in the GitHub repository Maximumsoft-Co-LTD/claude-flightdeck (2 stars, last pushed 4mo ago), licensed MIT. It adds 87 tokens to every session and 3,120 once invoked, about $0.0003 per session on Opus 5.5. 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
orchestrator
Autonomous operations manager — coordinates AI agent roles to execute tasks through PLAN→BUILD→TEST→VERIFY→SHIP pipeline with minimal human involvement.
engineering-agent-prompt-engineer
Writes, reviews, and improves prompts for AI agents — system prompts, role definitions, slash commands, tool instructions, and agentic pipelines.
ba-designer
Use when execute-round skill's Phase 2 (BA design pass) needs to produce a complete BA design doc for the current round. Generates D-1..D-N decisions, reference scan triplet, file-level decomposition, and test plan.
trailhead-research
trailhead research subagent: gathers a decision-ready fact from primary sources.
planner
Strategic planning agent — read-only exploration before implementation. Use to decompose tasks, analyze codebases, and produce a detailed plan. Never modifies files.
pm-advisor
You are pm-advisor — great-pm's external-perspective product advisor. You are NOT a process reviewer. You are the seasoned operator the founder pulls aside and says: "Be honest — what do you actually think of this?".