Write CFP (Call for Papers/Proposals) abstracts and speaker bios that get accepted. Covers hook sentences, problem framing, audience-fit statements, concrete takeaways, and adapting to different CFP formats (200/300/500 words). Use when submitting to any technical conference or meetup.
From clarcnpx claudepluginhub marvinrichter/clarc --plugin clarcThis skill uses the workspace's default tool permissions.
Designs and optimizes AI agent action spaces, tool definitions, observation formats, error recovery, and context for higher task completion rates.
Enables AI agents to execute x402 payments with per-task budgets, spending controls, and non-custodial wallets via MCP tools. Use when agents pay for APIs, services, or other agents.
Compares coding agents like Claude Code and Aider on custom YAML-defined codebase tasks using git worktrees, measuring pass rate, cost, time, and consistency.
A CFP abstract has seven sentences in order. Follow the structure exactly — reviewers scan dozens of abstracts and pattern-match on this structure.
Sentence 1: HOOK — a surprising statistic, provocative claim, or specific incident
Sentence 2: THE PROBLEM — what is broken, painful, or risky for the audience
Sentence 3: WHY NOW — what makes this problem urgent or newly solvable
Sentence 4: YOUR APPROACH — what you built, decided, or discovered
Sentence 5: THE EVIDENCE — what happened (metric, result, demo)
Sentences 6-7: TAKEAWAYS — three concrete things attendees will leave knowing or able to do
Sentence 8: AUDIENCE FIT — who this is for (role, experience level, prerequisite knowledge)
The average team merges a dependency with a known CVE every 6 days — without knowing it. As applications grow, manual security review of third-party packages becomes impossible at the velocity modern teams ship.
In 2023, we faced the same problem at Acme Corp: 800 direct dependencies, 40 engineers, and zero visibility into what changed. We built an automated dependency audit pipeline that classifies vulnerabilities by exploitability — not just severity — and integrates into every PR without slowing down review.
Attendees will leave knowing: (1) how to distinguish exploitable CVEs from theoretical risks using CVSS v3.1 vectors, (2) how to integrate Trivy or Grype into GitHub Actions in under an hour, and (3) a prioritization framework that reduces alert fatigue by 80%.
This talk is for engineers and security practitioners who manage CI/CD pipelines. Basic knowledge of GitHub Actions is helpful but not required.
| Reason | Example | Fix |
|---|---|---|
| Too vague | "We'll explore approaches to scaling" | Name the specific approach you took |
| No takeaways | Abstract describes what the talk is about, not what attendees gain | Add explicit "attendees will leave knowing X, Y, Z" |
| Vendor pitch | "Our product solves this problem perfectly" | Focus on the problem and technique, not the product |
| Audience unclear | No mention of who benefits | Add explicit audience fit sentence |
| Boring title | "Introduction to Docker" | Use a hook format (see below) |
| No evidence | "We improved performance significantly" | "We reduced P99 latency from 4.2s to 180ms" |
| Passive construction | "This talk will cover..." | "You will leave knowing..." / "We built..." |
The title is the first filter — reviewers decide in 2 seconds whether to read the abstract.
# How We [Did Something Specific and Surprising]
"How We Cut Deployment Time from 45 Minutes to 90 Seconds"
# [Number] Things You Need to Know About [Topic]
"5 Things Nobody Told You About Database Indexing"
# [Provocative Claim]
"Your Microservices Are Making You Slower"
"Most Observability Is a Lie"
# [Problem] Without [Sacrifice]
"Zero-Downtime Deployments Without Kubernetes"
"Test Coverage Without Slowing Down"
Write two versions for every submission: 50-word and 150-word.
Maria Chen is a Principal Engineer at Acme Corp where she leads the platform reliability team. She contributed to the OpenTelemetry specification and speaks regularly at KubeCon and SREcon. When not diagnosing distributed system failures, she runs ultramarathons.
Maria Chen is a Principal Engineer at Acme Corp, where she leads the platform reliability team responsible for 99.99% uptime across 200+ microservices. Before Acme, she spent four years at FinanceOps Inc rebuilding the settlement pipeline that processes $4B in daily transactions.
Maria contributed the baggage propagation specification to OpenTelemetry and has spoken at KubeCon NA, SREcon Americas, and QCon London. She writes at mariachen.dev about practical observability and the human side of on-call.
She is particularly interested in making reliability engineering accessible to teams without dedicated SRE functions — and has open-sourced a starter alerting template used by over 400 teams. When not diagnosing distributed system failures, she runs ultramarathons.
Adjust depth based on word count requirement:
**Section 1: The Hidden Cost of CVE Severity Scores (5 min)**
We examine why severity alone is a poor proxy for risk, using real examples
of high-severity CVEs that were not exploitable in practice.
**Section 2: Exploitability Vectors in CVSS v3.1 (10 min)**
...