Lengros

17 mods across 1 repository, 2 stars between them.

PreToolUse

01

Lengros/bennu

Hook Claude Code

Runs before the agent uses a tool for Write, Edit and NotebookEdit tool calls, executing guard.sh. From Lengros/bennu.

2 1mo ago A tokens not measured original MIT

Stop

02

Lengros/bennu

Hook Claude Code

Runs when the agent finishes a response, executing scan-secrets-stop.sh and learn-nudge.sh (2 commands). From Lengros/bennu.

2 1mo ago A tokens not measured copy · 52% MIT

SessionStart

03

Lengros/bennu

Hook Claude Code

Runs when a session starts, executing session-start.sh. From Lengros/bennu.

2 1mo ago A tokens not measured copy · 52% MIT

SessionEnd

04

Lengros/bennu

Hook Claude Code

Runs when a session ends, executing session-end.sh. From Lengros/bennu.

2 1mo ago A tokens not measured copy · 56% MIT

bennu

05

Lengros/bennu

Settings file Claude Code

Agent settings declaring 4 hook events (PreToolUse, Stop, SessionStart, SessionEnd).

2 1mo ago A tokens not measured copy · 81% MIT

design

06

Lengros/bennu

Skill Claude CodeCodex

Design a distinctive, heart-winning visual artifact — landing/marketing page, one-pager, hero, product page, pitch deck slide, theme — by first committing to a NAMED art direction, then hand-coding self-contained HTML/CSS, rendering it to a PNG via headless Chrome, and LOOKING at it to critique and fix. Use when the…

2 1mo ago A 229 tokens original MIT

diagram

07

Lengros/bennu

Skill Claude CodeCodex

Draw polished diagrams, schemas, architecture maps, data-flow / lifecycle diagrams, workflows, and infographics by rendering them from code (HTML/CSS + SVG → headless Chrome → PNG) — not by AI image generation. Use when the user wants a clear, editable visual of how a system, process, or set of relationships works, or…

2 1mo ago A 168 tokens original MIT

evidence-ledger

08

Lengros/bennu

Skill Claude CodeCodex

Maintain a living evidence ledger for a product bet — a portfolio of risky assumptions, each with a pre-registered success/kill threshold, an experiment matched to the exact question it can answer, the observed signal with provenance, and a forced kill/pivot/double-down decision. Use when validating a product idea…

2 1mo ago A 169 tokens original MIT

market-scan

09

Lengros/bennu

Skill Claude CodeCodex

Scan the public market for evidence on a product hypothesis — competitors and their pricing/coverage (supply lens) and what real users say about the pain in forums, app reviews, FB groups, YouTube/Reddit (demand lens). Spawns a fresh research subagent from a fixed brief template + source playbook, codes findings into…

2 1mo ago A 210 tokens original MIT

motion

10

Lengros/bennu

Skill Claude CodeCodex

How a UI element should MOVE — the craft layer of web motion. Decide whether a thing should animate at all, then pick the easing, duration, spring, origin, and interruptibility that make it feel right instead of generic. Use when building or polishing any interaction with motion — modals, drawers, toasts, dropdowns…

2 1mo ago A 227 tokens original MIT

prototype

11

Lengros/bennu

Skill Claude CodeCodex

Wire several screens into ONE self-contained, multi-screen, CLICKABLE HTML prototype the user opens in a browser and clicks through — no build, no server, no deploy, no framework. Its job is to validate a FLOW / interaction fast and one-shot a clickable mockup for a PM or design decision. Use when the user wants to…

2 1mo ago A 251 tokens original MIT

reframe

12

Lengros/bennu

Skill Claude CodeCodex

Turn a request into a verifiable problem statement before any work starts. Treats the literal ask as a proposed solution (output) and regresses it to the real result the user wants (outcome), then makes that outcome falsifiable. Use when a request arrives as an already-chosen action ("congratulate a teammate", "build…

2 1mo ago A 148 tokens original MIT

retro

13

Lengros/bennu

Skill Claude CodeCodex

Run when telemetry/retro.flag exists (RETRO DUE) or on the monthly scheduled cadence. Reads runs.jsonl + blocks.jsonl + lessons/INDEX.md, produces a 6-section friction/health report with proposals, and resets the retro trigger.

2 1mo ago A 54 tokens original MIT

review-motion

14

Lengros/bennu

Skill Claude CodeCodex

Review animation and motion code against a high craft bar — easing, timing, springs, origin/physicality, interruptibility, GPU-only properties, accessibility, asymmetric timing, cohesion. Default to flagging; approval is earned. Use when verifying any UI work that contains motion — a /prototype, production web…

2 1mo ago A 161 tokens original MIT

speech

15

Lengros/bennu

Skill Claude CodeCodex

Turn dictated raw ideas into a delivery-ready spoken script — capture dictation verbatim, structure it, translate if needed, and polish for the voice. Runs a two-persona pipeline (narrativewriter structures + translates → narrativeeditor polishes for speakability) against a fixed intake (language, audience, length…

2 1mo ago A 195 tokens original MIT

sprite-anim

16

Lengros/bennu

Skill Claude CodeCodex

Process and verify sprite/animation work for web UIs — normalize a set of animation frames to a uniform character size (area-scale + baseline-align) and verify CSS/JS animations in a real browser via the Chrome DevTools Protocol. Use when a task touches sprite frames, sprite-sheet alignment, or a decorative/UI…

2 1mo ago A 102 tokens original MIT

bennu CLAUDE.md

17

Lengros/bennu

Instructions file

Instructions for Lengros/bennu, covering bennu — operating system, 1. identity & loop, 2. startup, 3. intent — true task essence + concrete results and 4. cast — hire the best possible executor.

2 1mo ago A 2,677 tokens original MIT