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/GregorBiswanger/featherspecWrote 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/gregorbiswanger/featherspec/sdd-setup)<a href="https://agentmods.dev/commands/gregorbiswanger/featherspec/sdd-setup"><img src="https://agentmods.dev/badge/commands/gregorbiswanger/featherspec/sdd-setup/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/commands/gregorbiswanger/featherspec/sdd-setup"><img src="https://agentmods.dev/badge/commands/gregorbiswanger/featherspec/sdd-setup.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 | $0.00027 | $0.03717 |
| Opus 5.5 | $0.00011 | $0.01487 |
| Sonnet 5.5 | $0.00005 | $0.00743 |
| Haiku 4.5 | $0.00003 | $0.00372 |
Grade A, and why
sdd-setup 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 21d 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 — 249 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/sdd-setup — SDD Setup Wizard
The user may name a documentation language, project name, or stack after the command. If any are missing, ask for them in the wizard below.
You are running an onboarding wizard for this repository.
Session opening — language first, then orientation
Open under a 🪶 **FeatherSpec** banner with at most two short English sentences — a
warm one-line welcome to the template and that setup starts with one question — then
immediately ask Step 0 (the language question). Everything before the language is
chosen stays English and minimal: no SDD explanation, no command table yet.
A mixed-language opening is exactly what this ordering prevents. A re-run is a repository a
previous setup already touched — the Memory Bank carries real content instead of TBD
placeholders, or the snapshot has been reconciled at least once — or a run where the language
arrived as an argument. Only then open in that language, confirm it in one line instead of
re-asking Step 0, and compress the orientation to one sentence: a re-run tunes, it does not
re-onboard. The DocLanguage: value the template ships with is a default nobody chose, so it
never makes a run a re-run: a first setup asks Step 0 even though AGENTS.md carries a value.
After DocLanguage is set, give the whole orientation in DocLanguage:
- Explain Spec-Driven Development in 3–5 plain sentences, rendering this definition
faithfully in
DocLanguage(translate it — do not quote English into a non-English session): Spec-Driven Development (SDD) is an AI-assisted development approach where a clear, structured, versioned specification is the primary artifact. Implementation and tests are derived from the spec and continuously validated against it (generate → verify → refine). Because agents generate code anchored to an explicit spec — and their output can be checked, corrected and regenerated against it — SDD yields higher-quality, more maintainable code than unstructured "vibe" prompting. The spec survives the implementation and stays the reference point for change. - State this repo's operating protocol as
AGENTS.mddefines it at the top — the three-step form is load-bearing, so render the three steps inDocLanguagewithout adding or merging steps. - Show the
/sdd-*commands from the Commands table inAGENTS.md— all of them, inDocLanguage. That table is the single machine-facing roster; keep no second copy here. Name the usual order in one line, rendered from the Flow line under Commands.
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.
- 21d ago Changed · +31 lines e62fa4fe1bb3
- 27d ago Changed · +18 lines · +3 tokens per session 6eb48354a5cc
- 1mo ago First seen · 200 lines · 24 tokens per session scan A 67643a6ece44
sdd-setup is a command published in the GitHub repository GregorBiswanger/featherspec (68 stars, last pushed 20d ago), licensed MIT. It adds 27 tokens to every session and 3,717 once invoked, about $0.0001 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 2026-08-30.
Other commands, from other repositories
kiro-steering-custom
Command "kiro-steering-custom" from gotalab/cc-sdd, covering kiro custom steering creation, workflow, available templates, steering principles and tool guidance.
al-context-create
Generate or update context.md file documenting project structure, architecture, and key patterns for AI assistants and developers. ALDC workflow (Copilot prompt al-context.create); invoke explicitly.
plan
Run /plan — see .claude/commands/plan.md for the authoritative spec.
sddp-implement-qc-loop
Command description: Run implement and QC in a continuous loop. Argument hint: [optional: feature directory or branch name] Command category: orchestration Prerequisites: spec, plan, tasks.
sddp-prd
Command description: Create or refine the canonical product document. Argument hint: [rough product idea, users, domain, or market opportunity] Command category: project-bootstrap Prerequisites: none.
sddp-specify
Command description: Create a feature specification from a feature description. Argument hint: [feature description] Command category: feature-delivery Prerequisites: project-instructions.