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/trumb/claude-nginx-hardeningWrote 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/trumb/claude-nginx-hardening/manage-recipes)<a href="https://agentmods.dev/commands/trumb/claude-nginx-hardening/manage-recipes"><img src="https://agentmods.dev/badge/commands/trumb/claude-nginx-hardening/manage-recipes/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/trumb/claude-nginx-hardening/manage-recipes"><img src="https://agentmods.dev/badge/commands/trumb/claude-nginx-hardening/manage-recipes.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 | $0.00017 | $0.00970 |
| Opus 5 | $0.00009 | $0.00485 |
| Sonnet 5 | $0.00003 | $0.00194 |
| Haiku 4.5 | $0.00002 | $0.00097 |
Grade A, and why
manage-recipes scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context
- Saved recipes: !
ls learnings/recipes/*.md 2>/dev/null | wc -lrecipes found - Recent runs: !
ls -d outputs/run-* 2>/dev/null | wc -lruns in outputs
Operating Mode
DEFAULT: R0 + R1 (create, list, show, export) With run: escalates per recipe's execution_class With install: W1 (local) or X1 (remote)
Read @skills/nginx-hardening/RECIPES-SCHEMA.md for recipe format.
Subcommands
create
Interactive recipe builder:
- Name — Prompt for kebab-case name (validate: ^[a-z0-9]+(-[a-z0-9]+)*$)
- Description — What this recipe does
- Profile — Select: edge-public, internal-only, api-gateway, static-site, reverse-proxy-app, high-risk-lockdown
- Steps — Add steps one at a time:
analyze-logs— params: log paths, lookback days, include rotatedaudit-config— params: config targets, analysis level (1/2/3)check-ioc-feeds— params: feed names, advisory/stage modedeploy— params: target configs, sync repos, auto-pushcompact-learnings— params: threshold linessync-repos— params: repo paths
- Schedule — manual, cron expression, or systemd timer
- Auto-approve threshold — auto-accept findings at or below this severity (low, medium, or none)
- Confirmation checkpoints — which steps pause for approval
- Max privilege level — highest action class allowed (R0, R1, W1, W2, X1)
- Validate — Run schema-validator.py
- Save — Write to learnings/recipes/.md
run <recipe-name>
Execute a saved recipe:
- Load recipe from learnings/recipes/
- Verify all required_env_vars are set
- First execution: always show privilege summary and require confirmation
- Execute steps in order
- Pause at confirmation_checkpoints
- Auto-approve findings <= auto_approve severity threshold
- Write run results to outputs/run-/
list
Show all recipes: | Name | Description | Profile | Schedule | Last Run | Steps |
edit <recipe-name>
Load existing recipe, present current values, allow modification. Re-validate and save.
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 · 102 lines · 17 tokens per session scan A 1f2c07ef5e5b
manage-recipes is a command published in the GitHub repository trumb/claude-nginx-hardening (3 stars, last pushed 5mo ago), licensed MIT. It adds 17 tokens to every session and 970 once invoked, about $0.0001 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 2026-08-31.
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.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
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.