17 skills found
Cyptopimpinainteazy / 01-proof-before-claimNo completion claim is valid without proof from actual source, tests, wiring, and command output. - Before saying anything is "done", run the strongest proof command available.
benjam3n / agskAnalyze and construct arguments by checking logical validity, premise truth, hidden assumptions, and evidence strength. Read and execute `/home/ben/Documents/projects/reasoningtool/claude-code-plugin/skills/agsk/SKILL.md`, applying it to this input: $ARGUMENTS
Cyptopimpinainteazy / 02-no-fake-completionPrevent agents from using stubs, mocks, no-ops, or placeholder logic and calling them "implemented." - When you write a function body, it must contain real logic, not a stub.
Huaizz-shawen / check-statusOpenclaw + rosclaw + visual feedback
petar-nauka / fact-check-skillFact-checking and disinformation detection skill for Claude and Codex
bridge-mind / BridgeWardTrust nothing. Ship safely. — Skeptical-reading and prompt-injection defense skill for AI agents. Provenance tagging, red-flag patterns, refusal templates, and a read-only injection auditor. MIT.
benjam3n / cdValidate customer problems and solutions before building. Read and execute `/home/ben/Documents/projects/reasoningtool/claude-code-plugin/skills/cd/SKILL.md`, applying it to this input: $ARGUMENTS
netresearch / automated-assessment-skillSystematic project assessment against checkpoint-enabled skills with scripted verification + LLM review - by Netresearch
semaj90 / 20-validationlegal ai contextual chat
Themis128 / 10-ai-agent-rules- **Direct and technical:** Be direct and technical in responses — no conversational filler ("Great", "Certainly", "Okay", "Sure") - **One tool at a time:** Wait for user confirmation after each tool use before proceeding - **Verify before completing:** Check task requirements, verify output files e…
carlosamesar / contract-uiDefinir la interfaz estable entre la experiencia de usuario y las fuentes de datos: props, estados, acciones, eventos, mensajes de error y expectativas de renderizado. ---
Cyptopimpinainteazy / 05-tests-are-sacredTests are the proof mechanism. Modifying tests to hide code failures is sabotage. Failing tests signal real problems. - Run the full test suite for the changed area, not just the file you touched.
DangQuangSE / Inversion ExerciseFlip core assumptions to reveal hidden constraints and alternative approaches - "what if the opposite were true?"
opslane / opslaneVerification Layer for Claude Code
benjam3n / stEvery strategy exists within a game (competitive context). Every game. Read and execute `/home/ben/Documents/projects/reasoningtool/claude-code-plugin/skills/st/SKILL.md`, applying it to this input: $ARGUMENTS
zhjai / agent-arenaEvidence-first multi-agent debate skill: get a second opinion by pitting Codex × Claude Code (or GLM/DeepSeek/Qwen) to independently review, red-team & judge high-stakes code and architecture decisions.
PlaiPin / check-statusROS2 meets OpenClaw 💗
