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 agentmods add commands/arome3/code-to-content/cascadegit clone --depth 1 https://github.com/arome3/code-to-contentWhat 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 | $0.00000 | $0.01893 |
| Opus 5 | $0.00000 | $0.00946 |
| Sonnet 5 | $0.00000 | $0.00379 |
| Haiku 4.5 | $0.00000 | $0.00189 |
Grade A, and why
cascade 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 2d 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 — 306 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/c2c:cascade
Automatically derive multiple content formats from a single source piece. Write once, publish everywhere.
Usage: /c2c:cascade <source-file> [--formats twitter,linkedin,newsletter]
Differentiation: Repurposing multiplies reach, but a generic source piece stays generic on every platform. Make sure the source leads with the WHY, lands one defensible opinion, and shows a road not taken before cascading. Run the swap-the-name test from
references/differentiation.mdon each adapted piece.
What Cascade Does
Takes your primary content (blog post, tutorial, etc.) and automatically generates platform-optimized versions:
Blog Post (source)
│
├──► Twitter Thread (key insights, 8-12 tweets)
├──► LinkedIn Post (professional angle, 800-1300 chars)
├──► Newsletter Section (subscriber value-add)
└──► README Section (if applicable)
All derived content:
- Maintains voice consistency with source
- Extracts the most compelling insights
- Adapts depth for each platform
- Respects platform-specific constraints
Reference Loading
Load these references for cascade:
references/phase-gates.md(always)references/social-content.mdreferences/formats.md(sections for target formats)references/checklists.md(sections for target formats)
Cascade Workflow
Step 1: Analyze Source Content
Read the source file and extract:
- Core insight — The main takeaway
- Key points — 3-5 supporting ideas
- Evidence — Code snippets, metrics, examples
- Voice profile — Tone and style used
- Audience level — Beginner/intermediate/expert
## Source Analysis
**Core Insight:** [One sentence summary]
**Key Points:**
1. [Point with evidence]
2. [Point with evidence]
3. [Point with evidence]
**Voice:** [detected voice profile]
**Audience:** [detected level]
Step 2: User Confirmation Checkpoint
PAUSE — Ask user to confirm before generating:
I've analyzed your [format]: "[title]"
Core insight: [insight]
Key points: [3-5 points]
Voice: [voice profile]
I'll generate these formats:
- Twitter thread (8-12 tweets)
- LinkedIn post (~1000 chars)
- Newsletter section
Proceed with cascade? [Yes / Modify formats / Cancel]
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.
- 2d ago First seen · 306 lines · 0 tokens per session scan A 78a4704819c2
cascade is a command published in the GitHub repository arome3/code-to-content (6 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,893 tokens. 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
retrospective
Perform a post-implementation retrospective analysis measuring spec adherence, implementation deviations, and lessons learned.
protracted-strategy
用持久战略为长期复杂任务划分阶段、设置转折条件。.
resume_handoff
Resume work from handoff document with context analysis and validation.
distill
Distill important session insights into doc/loom/knowledge.
demo-command
Example slash command that wraps the demo-skill. Showcases the command kind.
speckit.verify-tasks
Verify tasks marked [X] in tasks.md are implemented, not phantom completions (marked done but backed by missing or dead code).