hook-everything
AGENTS.md19 starsChanged 9 days agoMITView on GitHub
Read this file first in every session.
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository14 other files it gives its agents.
- hook-everythingSkill
- anti-slopSkill
- audience-calibrationSkill
- creative-ideationSkill
- hook-compoundingSkill
- marketing-hooksSkill
- market-intelligenceSkill
- product-hooksSkill
- psychology-mapSkill
- self-reviewSkill
- trust-architectureSkill
- value-checkSkill
- visual-ux-hooksSkill
- writing-hooksSkill
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 3
Read this file first in every session.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AGENTS.md
Read this file first in every session.
Core Operating Principles
- Live Market Grounding with the Organic Fit Gate requires loading
skills/market-intelligencebefore ideating campaigns or positioning. Run live web searches to capture real Voice-of-Customer (VoC) phrases and competitor blind spots. Pass every market signal through the Organic Fit Gate so trends are woven in only when causally connected to the user's problem. Drop any trend that feels forced. - Audience Cognitive Routing uses
skills/audience-calibration(Petty & Cacioppo's ELM + Construal Level Theory) to match hook velocity, tone, and proof type to the exact reader (Practitioner/Developer vs. Executive vs. Consumer vs. Peer Forum). - Trust Before Any Hook means pairing every curiosity or conversion hook with a Trust Damper from
skills/trust-architecture(Pratfall admission, Inoculation, Operational Transparency, or Benevolent Anti-Pitch). - 4-Note Psychological Chords come from
skills/hook-compounding. Layer 3 to 4 complementary mechanisms at once (Trust Damper + Cognitive Anchor + Active Generator + Forward Vector), and pair text with spatial structure fromskills/visual-ux-hooks. - Anti-Scam High Conversion follows
skills/marketing-hooksto lift conversion via Self-Perception micro-commitments (Bem 1967), Mental Accounting reframes (Thaler 1985), and Regret Regulation (Zeelenberg 1999), never via fake urgency or inflated ROI. - 85-Rule Deep Anti-Slop Across Languages requires every draft to obey
skills/anti-slopand score<= 15/100onpython3 tools/slop_scan.py <file> --fix-hints. Rewrite in 300-word blocks to prevent LLM voice drift. - Self-Review in Any Language runs
skills/self-reviewafter the scan: reread the draft as a native reader of its language and rewrite whatever still sounds machine-made, up to 3 rounds. - Mandatory 13-Point Value Gate runs
skills/value-checkbefore delivering output.
Lite Lane vs. Full Lane Routing
- Lite Lane (Fast Path for Short Edits & Single Pieces) applies when the user asks to humanize existing text, write a short email or release note, or tighten a single headline. Load at most 3-4 skills (
writing-hooksorproduct-hooks->anti-slop->self-review->value-check) and runtools/slop_scan.py. - Full Lane (Deep Architecture for Campaigns, Funnels & New Products) applies when building landing pages, cold outreach sequences, onboarding flows, or trend-grounded ideation. Follow the full routing table below.
Skill Routing Table
| Task / Domain | Required Skills (Load in Order) |
| Market research, positioning, competitor teardown, or trend analysis | market-intelligence -> audience-calibration -> trust-architecture -> value-check |
| Brainstorming fresh hook ideas, content angles, or campaign concepts | market-intelligence -> creative-ideation -> hook-compounding -> anti-slop -> self-review -> value-check |
| Marketing campaigns, landing pages, pricing, cold outreach, ads, launches | market-intelligence -> audience-calibration -> trust-architecture -> hook-compounding -> marketing-hooks -> visual-ux-hooks -> anti-slop -> self-review -> value-check |
| Essays, articles, newsletters, scripts, documentation, copy | audience-calibration -> trust-architecture -> hook-compounding -> writing-hooks -> anti-slop -> self-review -> value-check |
| Product onboarding, SaaS flows, UI/UX layouts, retention, projects | psychology-map -> trust-architecture -> visual-ux-hooks -> hook-compounding -> product-hooks -> value-check |
| Rewriting or humanizing text in any language (Lite Lane) | anti-slop (plus tools/slop_scan.py) -> self-review -> value-check |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.