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 provos/ironcurtain --skill summary-style-guidegit clone --depth 1 https://github.com/provos/ironcurtainWrote 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/provos/ironcurtain/summary-style-guide)<a href="https://agentmods.dev/skills/provos/ironcurtain/summary-style-guide"><img src="https://agentmods.dev/badge/skills/provos/ironcurtain/summary-style-guide/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/provos/ironcurtain/summary-style-guide"><img src="https://agentmods.dev/badge/skills/provos/ironcurtain/summary-style-guide.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 | $0.00048 | $0.00460 |
| Opus 5 | $0.00024 | $0.00230 |
| Sonnet 5 | $0.00010 | $0.00092 |
| Haiku 4.5 | $0.00005 | $0.00046 |
Grade A, and why
summary-style-guide 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.
What it actually says
Summary style guide
Each email in the report gets a one- or two-sentence prose summary. This document covers the voice of those sentences. For the surrounding markdown structure, the heading depth, and where the summary sits in the document, see the email-formatting conventions.
Voice and tense
- Third-person, past tense. "Alice asked the team to review the deck" rather than "I am asking..." or "Alice is asking...".
- Drop the salutation, the signature, and any thread quoting. The summary is what the email did, not how it was framed.
- Active voice. "Bob declined the meeting" not "the meeting was declined by Bob".
What to include
The summary should answer: what is the actionable substance of this email? Concretely:
- The ask, decision, or status update at the heart of the message.
- Who is on the hook for what, if explicit.
- Any deadline or date that anchors the action.
What to omit
- Niceties ("Thanks!", "Hope you're well", "Cheers").
- The sender's restatement of what they previously said in the thread.
- Speculation about why the email was sent.
- Any attachment list — that's separate metadata, not summary content.
Length
- Prefer one sentence. Use two when the email genuinely covers two distinct items (e.g., a status update and an unrelated ask).
- Hard cap: two sentences. If a summary feels like it needs three, the email contains multiple threads of substance and the summary should pick the most actionable one and drop the rest.
Confidence and hedging
If the email is ambiguous or the body is truncated, write a confident summary of what's visible rather than hedging. Do not insert "appears to" or "seems to be" — the consumer of the summary already knows it was machine-derived.
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 · 52 lines · 48 tokens per session scan A e4cd03268a7a
summary-style-guide is a skill published in the GitHub repository provos/ironcurtain (602 stars, last pushed 5d ago), licensed Apache-2.0. It adds 48 tokens to every session and 460 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 skills, from other repositories
09-for-sure
Run an iterative agent loop that retries until a runnable success condition passes. Use when the user says "for sure", "keep trying until", or wants guaranteed completion against a success command. Not for one-shot tasks or uncheckable goals.
maintenance
Update mcp-reporter dependencies, adopt SDK changes, reconcile development workflows, and verify the CLI and library.
field-test
Exercise mcp-reporter's collected reports through in-memory, stdio, HTTP and compiled Node entry points.
ci-cd-pipeline
Design and implement CI/CD pipelines for automated testing and deployment.
test-driven-development
TDD workflow — Red, Green, Refactor cycle with practical examples.
resume-ai-validator
A command-line tool for testing ResumeAI, a service that analyzes and improves résumés. It checks both the service's API and the user interface through complete user journeys.