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/juanklagos/spec-driven-development-template/newgit clone --depth 1 https://github.com/juanklagos/spec-driven-development-templateWhat 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 | $0.00032 | $0.00574 |
| Opus 5 | $0.00016 | $0.00287 |
| Sonnet 5 | $0.00006 | $0.00115 |
| Haiku 4.5 | $0.00003 | $0.00057 |
Grade A, and why
new 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.
What it actually says
Guide the user from zero to a first spec ready for approval. Respond in the user's language (EN/ES). Ask ONE short question at a time; no jargon.
- If
$ARGUMENTSis empty, ask for the project in plain language. Then run a short discovery interview — one question at a time, only what is missing:- Problem and main goal (who hurts today, what changes when this works?).
- MVP scope: what is IN and what is explicitly OUT.
- Constraints: deadline, budget, tech the user must/can't use, existing systems to respect.
- Stack: only if the user has one in mind; otherwise propose later in
plan.md, not here. - Edge cases and risks: "what's the worst realistic thing a user could do or expect?" Summarize the interview back in 5 bullets and get a "yes" before writing anything.
- Decide the layout with the user:
- Existing/external project → compact sidecar:
./scripts/install-spec-sidecar.sh <path> --profile=recommended(SDD artifacts in./spec/, code stays in the project root). - Project living inside this template repo →
./scripts/create-www-project.sh <name> <agent>.
- Existing/external project → compact sidecar:
- Fill
idea/IDEA_GENERAL.md(or./spec/idea/IDEA_GENERAL.md) with name, problem, goal, MVP scope. - Create the first spec bundle:
./scripts/new-spec.sh "<feature>" "<owner>"(sidecar:./spec/scripts/new-spec.sh). Visual alternative: the SDD Builder can create and connect spec cards on a canvas (guide 51). / Alternativa visual: el SDD Builder crea y conecta tarjetas de spec en un lienzo (guía 51). Its ✨ assistant drafts the whole board (idea → epics → specs) from one sentence. / Su asistente ✨ genera el board completo (idea → épicas → specs) desde una frase. - Draft
spec.mdwith the user: user story, Given/When/Then scenarios, EARS acceptance criteria (WHEN [trigger], THE SYSTEM SHALL [behavior]), out of scope. - Ask the user to approve the spec; record status/date/approver/evidence in its approval section.
- Close by pointing to
/sdd:gateas the next step.
Hard stop: do NOT write project code in this command. Base creation needs no consent; implementation later will.
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 · 25 lines · 32 tokens per session scan A a5eef6fe8d29
new is a command published in the GitHub repository juanklagos/spec-driven-development-template (15 stars, last pushed 2d ago), licensed MIT. It adds 32 tokens to every session and 574 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 2026-08-30.
Other commands, from other repositories
paul:help
Show available PAUL commands and usage guide.
test
Smart test runner with filtering, coverage, and health monitoring.
safe-refactor
Safe refactoring with automated review, testing, and rollback capabilities.
expect
Diff-aware AI browser testing — reads the git diff, maps changes to affected pages via the route map, generates a targeted test plan, and executes it via agent-browser (Rust daemon + CDP, ARIA-tree-first) with pass/fail reporting. Use when testing UI changes, verifying PRs before merge, or running regression checks on…
paul:verify
Guide manual user acceptance testing of recently built features.
drift
Post-implementation spec drift check — verify the implementation matches existing OpenSpec specifications.