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 skills add glebis/claude-skills --skill feature-factorygit clone --depth 1 https://github.com/glebis/claude-skillsWrote 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/skills/glebis/claude-skills/feature-factory)<a href="https://agentmods.dev/skills/glebis/claude-skills/feature-factory"><img src="https://agentmods.dev/badge/skills/glebis/claude-skills/feature-factory/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/skills/glebis/claude-skills/feature-factory"><img src="https://agentmods.dev/badge/skills/glebis/claude-skills/feature-factory.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00124 | $0.03059 |
| Opus 5 | $0.00062 | $0.01529 |
| Sonnet 5 | $0.00025 | $0.00612 |
| Haiku 4.5 | $0.00012 | $0.00306 |
Grade A, and why
feature-factory 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 11d 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Feature Factory
A goal-driven, local-first loop for taking one feature from intent to shipped. The whole method exists to hold a single line in tension: don't let the process outrun the feature. Keep the spine (goal → TDD → deterministic verify → human merge → evidence-when-it-matters); delete ceremony aggressively.
This is a behavior guide, not an engine. Do not build generic config, executors, telemetry, optimizers, or a universal factory verify wrapper. Run the behavior; package nothing the feature didn't earn.
Core principle
The human defines the desired system state. Agents maintain the specifications. Tests and evidence decide whether reality complied. Two human gates are always required — approve the Goal Contract (cheap-to-change moment) and review the merge (irreversible moment) — plus a conditional third (plan approval) when a size/risk trigger fires (see step 2). Everything between is a single focused agent loop.
When to apply vs skip
- Apply to a bounded, shippable feature.
- Skip the heavy parts for trivial changes — an S-size fix is just: short goal in your head → TDD → run the repo's checks → merge. Don't generate documents for a one-liner.
- Refuse XL — if the feature is multi-day with shared contracts, migrations, auth/billing, or product ambiguity, split it first; do not run an XL feature through this loop whole.
The loop (six steps)
1. Intake — create or repair a Goal Contract
Make a quick size call now (step 2 formalizes it) — you need it to decide how heavy intake should be. For S-size/trivial changes, skip the file — a short goal stated in chat and confirmed by the human is enough; jump to step 3. Otherwise, copy assets/goal-contract.md into the target repo as goal.md under a feature dir (suggested: docs/factory/<date>-<slug>/goal.md); see assets/goal-contract-example.md for a filled example of the calibration expected. The template's core fields are enough for M features — the conditional half is for L or when a trigger fires. Draft it from the request first, then have the human confirm/correct each field — don't block on a blank form, and don't proceed past intake until the human has approved the wording. Enforce:
- All
<!-- required -->fields present: Smallest shippable slice and Stop condition. - Respect the caps (
≤3,≤5). A capped goal stays a goal, not waterfall-in-markdown. - Every desired outcome maps to concrete evidence. Reject vague/solution-coupled outcomes.
- Fail rule: if a goal can't produce evidence, it's a wish with better formatting — it doesn't pass.
- Agents may propose Goal Amendments; never silently rewrite the goal.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 11d ago First seen · 93 lines · 124 tokens per session scan A 078eb5525bbf
feature-factory is a skill published in the GitHub repository glebis/claude-skills (374 stars, last pushed 8d ago), licensed MIT. It adds 124 tokens to every session and 3,059 once invoked, about $0.0006 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 skills, from other repositories
engram-testing-coverage
TDD and coverage standards for Engram. Trigger: When implementing behavior changes in any package.
iterative-development
TDD iteration loops using Claude Code Stop hooks - runs tests after each response, feeds failures back automatically.
python
Python development with ruff, mypy, pytest - TDD and type safety.
nw-fp-clojure
Clojure language-specific patterns, data-first modeling, REPL-driven development, and spec.
strict-tdd
Strict RED->GREEN->REFACTOR test-driven development with enforcement. Never write production code before a failing test. Atomic commits per TDD cycle.
mobiai-mobile-tdd
You MUST use this before writing any implementation code for a mobile feature, bug fix, refactor, or behavior change. Tests come before implementation — no exceptions.