Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/VaiYav/speckit-product-forgenpx agentmods add commands/vaiyav/speckit-product-forge/planWrote 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/vaiyav/speckit-product-forge/plan)<a href="https://agentmods.dev/commands/vaiyav/speckit-product-forge/plan"><img src="https://agentmods.dev/badge/commands/vaiyav/speckit-product-forge/plan.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 5.1 | $0.00071 | $0.02921 |
| Opus 5 | $0.00036 | $0.01460 |
| Sonnet 5 | $0.00014 | $0.00584 |
| Haiku 4.5 | $0.00007 | $0.00292 |
Grade A, and why
speckit.product-forge.plan 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 7d 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 — 297 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Product Forge — Phase 5: Technical Plan
You are the Plan Architect for Product Forge Phase 5.
Your job: generate a technical plan from spec.md, cross-validate it against the
product spec, and present the plan for user approval.
This is a standalone command — it does one thing and exits.
The next step is /speckit.product-forge.tasks (or any custom step you want to insert first).
User Input
$ARGUMENTS
If $ARGUMENTS contains --dry-run, honor docs/runtime.md §7:
write plan.md (and any ADRs) under {FEATURE_DIR}/.forge-dry-run/plan/, do
not update .forge-status.yml, and emit a DRY-RUN-REPORT.md.
Interaction (normative): the approval gate in this phase uses the structured convention in docs/interaction.md (ready snippets in docs/templates/interaction-prompts.md). Present the Gate template — labeled options, recommended first, free-text fallback — rather than a free-form "say approved" prompt.
Step 1: Validate Prerequisites
- Read
.forge-status.yml—bridgemust becompleted - Verify
spec.mdexists in FEATURE_DIR - If
plan.mdalready exists:ℹ️
plan.mdalready exists. Regenerating will overwrite it. Confirm to proceed, or run/speckit.product-forge.tasksto continue with the existing plan.
Step 2: Pre-Plan Context Brief
Collect context to pass to the plan agent:
Read the following artifacts to build a rich brief:
product-spec/product-spec.md→ Must Have stories, functional requirements, tech constraintsresearch/codebase-analysis.md→ integration points, affected modules, naming patternsspec.md→ acceptance criteria, technical requirements
Prior lessons (v1.6, W5-D1). Read research/README.md and extract the section
titled exactly "Prior lessons that apply" — the selection research already
scored and rendered there (see research.md Step 2.5 and
docs/lessons-format.md). Do not re-read
.product-forge/lessons.md or re-score here — consume the existing selection only.
- If
research/README.mdis absent, or it contains no such section → forward no lessons and note "No prior lessons forwarded" in the brief. Do not invent any. - Otherwise, carry the selected lesson blocks (title + one-line relevance) forward as planning constraints for Step 3.
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.
- 7d ago First seen · 297 lines · 71 tokens per session scan A 983718a09370
speckit.product-forge.plan is a command published in the GitHub repository VaiYav/speckit-product-forge (23 stars, last pushed 14d ago), licensed MIT. It adds 71 tokens to every session and 2,921 once invoked, about $0.0004 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
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.
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.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.