docs(front): reframe landing around the acceptance-layer narrative#248
Conversation
Restructure the front page flow after the h5i serve demo: add a 'What you just saw' interpretation, sharpen the problem into 'code generation got cheap, code acceptance got risky', collapse the 6-card feature grid into three primitives (Isolate / Record / Audit & Accept), and replace the tricks section with a 'What lives in Git' refs/h5i/* section. FAQ now leads with 'What is h5i?' to broaden past team framing.
… caption Drop the standalone acceptance-layer section and instead put a single line under the h5i serve demo: 'A live record of what each agent did, why it changed the code, and what evidence supports merging it.'
Insert a quick-scan section after the hero: the README elevator-pitch paragraph plus a six-card capability grid (prompt versioning, persistent context/memory, supervised sandbox, ~95% token reduction, conflict-free multi-agent orchestra, automated audit) so readers grasp the purpose before the 'bottleneck changed' framing.
- Add a top border/padding to #team so there is a divider line between the 'How it works' primitives and the team section. - Remove the bottom 'Get Started' install section; the /workflows guide and the hero install command already cover onboarding. - Repoint the nav 'Install' link to the hero install command (#quickstart) at the top of the page, and drop the redundant hero 'Get Started' button, promoting 'See Workflows' to the primary action.
Relocate the 'two heads are better than one' proverb row from the 'Built with h5i' section to the end of the killer-workflow team section, where the 'run many, merge one' framing makes it land.
Swap the split terminal transcript + verdict cards in the team section for the _static/hero-team.svg diagram, framed in a dashboard-wrap.
🪙 75% AI-authored · 1 file · prompt 51/100
Important 🎯 Goal: Restructure docs front page: acceptance-layer narrative, 3 primitives (Isolate/Record/Audit & Accept), dissolve tricks section, tighten FAQ
💬 The ask
📍 What shipped
Note 🌿 Prompt maturity: 51/100 · proficient · 6 prompts scored (100% of AI commits) 📊 heuristic breakdown
Readability (display-only): Flesch 60 · FK grade 9.8 · Gunning Fog 12.3 Offline classical-NLP score — no LLM. Higher = more specific, bounded, and grounded prompting. Tip ✅ Security scan cleanNo credentials leaked in this branch's diff. h5i scanned every added line against the secret rule pack and found nothing to rotate. Note Duplicate-code scan clean — no copy-paste blocks introduced. 🧠 Reasoning by file — 10295 nodes across 2 lanes, latest 24 only%%{init: {"flowchart": {"nodeSpacing": 42, "rankSpacing": 48, "diagramPadding": 14}, "themeVariables": {"fontSize": "18px"}} }%%
flowchart TB
subgraph lane_0["💭 Reasoning · 3 steps"]
direction LR
n_lane_0_0["THINK · Let me do a quick structural sanity check on tag balance:"]:::t
n_lane_0_1["THINK · Let me add a compact ”What h5i is” section right after the hero (before …"]:::t
n_lane_0_2["THINK · I'll move the proverbs block from `#built-with` into the `#team` section…"]:::t
n_lane_0_0 --> n_lane_0_1
n_lane_0_1 --> n_lane_0_2
end
subgraph lane_1["📄 docs/index.html · 21 ops"]
direction LR
n_lane_1_0["READ"]:::o
n_lane_1_1["EDIT × 7"]:::a
n_lane_1_2["READ"]:::o
n_lane_1_3["EDIT × 7"]:::a
n_lane_1_4["READ"]:::o
n_lane_1_5["EDIT × 2"]:::a
n_lane_1_6["READ"]:::o
n_lane_1_7["EDIT"]:::a
n_lane_1_0 --> n_lane_1_1
n_lane_1_1 --> n_lane_1_2
n_lane_1_2 --> n_lane_1_3
n_lane_1_3 --> n_lane_1_4
n_lane_1_4 --> n_lane_1_5
n_lane_1_5 --> n_lane_1_6
n_lane_1_6 --> n_lane_1_7
end
classDef o fill:#dbeafe,stroke:#1e3a8a,color:#0b1c4a;
classDef t fill:#fef3c7,stroke:#92400e,color:#3f2d05;
classDef a fill:#dcfce7,stroke:#166534,color:#0a2e16;
classDef n fill:#ede9fe,stroke:#5b21b6,color:#221251;
classDef m fill:#e5e7eb,stroke:#374151,color:#0b0f17;
📜 Per-commit provenance — 6 AI-authored commits
|
No description provided.