Skill Claude CodeOpenCode
Autonomous state machine that drives a task through its full lifecycle (research -> plan -> implement -> complete) without user confirmation between phases. Invoke for /orchestrate command.
Skill Claude CodeOpenCode
Autonomous state machine that drives a task through its full lifecycle (research -> plan -> implement -> complete) without user confirmation between phases. Invoke for /orchestrate command.
Agent Claude Code
Implement general, meta, and markdown tasks from plans with hard-mode behavioral contracts.
Agent Claude Code
Research general tasks using web search and codebase exploration with hard-mode behavioral contracts.
Agent Claude Code
Create phased implementation plans with hard-mode behavioral contracts for complex, deflection-prone tasks.
Agent Claude Code
Multi-output synthesis for team skills. Reads all teammate finding files in its own fresh context, resolves conflicts, identifies gaps, and writes a unified research report.
Command Claude Code
Fetch a Zulip thread via API and write formatted JSON to a file.
Skill Claude Code
Execute hard-mode implementation with anti-analysis contracts, per-phase dispatch, and territory-aware execution. Invoke for --hard implementation tasks.
Skill Claude Code
Full structural hard-mode orchestration state machine with per-phase dispatch (H1), adversarial verification (H4), convergence policing (H6), territory contracts (H7), and churn detection (H5). Invoke for /orchestrate --hard.
Skill Claude Code
Create hard-mode implementation plans with phase sizing, postmortem constraints, and preserved-assets accounting. Invoke for --hard planning tasks.
Skill Claude Code
Conduct hard-mode research with adversarial verification, reference grounding, and anti-analysis contracts. Invoke for --hard research tasks.
Skill Claude Code
Fetch a Zulip thread via API and write formatted JSON to a file. Invoke for /zulip command.
Agent Claude Code
Implement CSLib proofs following Lean 4 and CSLib contribution standards.
Agent Claude Code
Implement CSLib proofs with hard-mode behavioral contracts (H2 anti-analysis, H7 territory, H9 wrap-up discipline).
Agent Claude Code
Research CSLib formalization patterns and Mathlib API for CSLib contributions.
Agent Claude Code
Research CSLib formalization patterns and Mathlib API with hard-mode behavioral contracts (H2 anti-analysis, H3 reference grounding, H4 adversarial verification).
Agent Claude Code
Vet CSLib tasks against library standards, run CI pipeline, and create fix tasks with interactive user confirmation.
Agent Claude Code
Read PR review research and compose pr-response.md and/or zulip-response.md for pr-type review tasks.
Agent Claude Code
Fetch and synthesize GitHub PR review discussion and Zulip thread content for pr-type review tasks.
Command Claude Code
Create and submit a CSLib PR, or create a PR review task (--review).
Command Claude Code
Vet completed CSLib tasks against library standards (CONTRIBUTING.md, NOTATION.md, ORGANISATION.md, CODEOFCONDUCT.md), run the CI pipeline, and create fix tasks for violations found.
Skill Claude Code
Implement CSLib proofs with hard-mode contracts (H2 anti-analysis, H7 territory, H9 wrap-up with sorryinventory). Invoke for --hard cslib implementation tasks.
Skill Claude Code
Implement CSLib proofs following Lean 4 and CSLib contribution standards. Invoke for cslib implementation tasks.
Skill Claude Code
Research CSLib formalization patterns with hard-mode contracts (H2 anti-analysis, H3 reference grounding with BibKey verification, H4 adversarial verification). Invoke for --hard cslib research tasks.
Skill Claude Code
Research CSLib formalization patterns and Mathlib API for CSLib contributions. Invoke for cslib research tasks.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: