0SxD

@0sxd source repo

19 skills
0 followers
6 hours ago last updated

▌
Ingest · 0sxd bundle
Use this skill when absorbing an external project bundle, prompt pack, agent packet, or skill folder into a curated repository, especially when the source needs to be evaluated for publish-readiness before it lands in the canonical tree. Triggers: any request to ingest a packet, absorb a folder, evaluate a prompt pack against a rubric, route a third-party skill through gates, or convert a NotebookLM source bundle into a SKILL.md-conformant skill. The skill runs an interrogation loop with the human if the source is incomplete, never assumes missing fields, and produces a quarantine-or-promote verdict at the end. Do NOT use this skill for editing skills already in the canonical tree (use github-curator), for evaluating finished artifacts ready to publish (use publish-readiness-rubric), or for arbitrary file ingestion unrelated to the Agent Skills format.

0
installs
▌
Github Curator · 0sxd bundle
Use this skill when curating, evolving, or publishing a GitHub repository that contains Anthropic Agent Skills, especially when the repo must remain self-consistent across edits. Triggers: any request to add a skill, edit a skill, bump a version, cut a release, audit the repo against the Agent Skills spec at agentskills.io, validate AGENTS.md and CLAUDE.md alignment, run the four-gate publish-readiness check on this repo itself, or onboard a new contributor to the curator pattern. The skill operates on its own repository as the first example: every action it instructs the agent to take, it has already taken on this repo. Do NOT use this skill for arbitrary GitHub tasks unrelated to skill curation, for non-Skills repos, or when the user has asked for code-level changes to a separate codebase.

0
installs
▌
Session Dispatch · 0sxd bundle
Use this skill when spawning a new chat session for a focused workstream that exceeds the current session's context budget, especially when the new session must run in any agent (Claude Code, Codex, Cursor, Aider, Gemini CLI) or any host (Claude.ai project, NotebookLM, custom Gem, custom GPT). Triggers: any request to write a handoff prompt, dispatch a workstream, spawn a sub-agent session, prepare a paste-ready bootstrap for a new chat, or generate a HANDOFF document at 75% context. The skill emits a single AGENTS.md-shaped bootstrap file plus an optional packet zip that the human pastes or drops into the new session. The handoff is host-neutral: the same file works in every supported environment because it conforms to the AGENTS.md standard. Do NOT use this skill for in-session task delegation (use the agent's own subagent feature), for resuming an existing session, or for arbitrary chat templates unrelated to workstream dispatch.

0
installs
▌
Publish Readiness Rubric · 0sxd bundle
Use this skill when scoring an artifact's readiness for publication on GitHub, especially when the artifact must pass a deterministic boolean-atomic check before it ships. Triggers: any request to evaluate an artifact for publish-readiness, run the four-gate check, score against the Trinity Dialectic 3x3, generate a publish-readiness report, or apply criterion tags (Essential / Important / Optional / Pitfall) to a structured evaluation. The skill returns a binary verdict per row plus an aggregate verdict (PROMOTE / QUARANTINE / PROMOTE-WITH-WARNING). The rubric is mechanical, deterministic, and reproducible: the same input produces the same output. Do NOT use this skill for subjective quality assessment, for free-form code review, or as a substitute for the human's Pathos-Ethos-Logos verdict (PEL=1 is required and cannot be self-issued by the agent).

0
installs
▌
Setup Intake · 0sxd
Use this skill immediately after system_directive at session start. Captures the irreducible parameters needed to run any CE_RD_OS workstream: mission, role, sources, rubric mode, sub_agent profile, memory option, ML backend option, loop bound, output target. Trigger when the user says "set up", "let's begin", "what do you need from me to start", or when the agent recognizes a new workstream is being initiated. Asks at most three questions per turn.

0
installs
▌
Publish Bundle · 0sxd
Use this skill to package a CE_RD_OS workstream output into the dual-track delivery format: GitHub repo plus 10-file text zip. Trigger when the user says "ship it", "package this", "publish v0.X.Y", "make me a release", "pack the bundle", or when a workstream reaches its termination criteria. Runs verify_self.sh as the privacy and conformance gate, runs pack_text_bundle.sh to produce the zip, fills the approval_gate template honestly, and surfaces to the architect for verdict. Never self-issues the release verdict.

0
installs
▌
Trinity Rubric · 0sxd
Use this skill whenever an output, a source, or a plan needs to be evaluated against the project's quality bar. Trigger when the agent finishes a draft and is about to surface it, when the user says "score this", "review this", "is this good enough to ship", or when the publish_bundle skill runs its publish-readiness check. The 9-cell Pathos / Ethos / Logos by Coverage / Self-contained / Importance grid grounded in arXiv:2507.17746. Default mode boolean-atomic. Optional ML backend hook for GEPA via MLflow or OpenEvolve.

0
installs
▌
Source Ingestion · 0sxd
Use this skill whenever new source material enters the workstream: uploaded files, URLs, paths to local folders, MCP-attached datasets, or pasted text blocks. Trigger when the user says "ingest this", "absorb these sources", "add to the corpus", "read this and apply", or attaches files at session start. Runs the DETECT, NORMALIZE, INTERROGATE, SCORE, ROUTE loop. Quarantines anything that cannot be conformed.

0
installs
▌
System Directive · 0sxd
Use this skill at session start, before any other skill. Establishes identity, source hierarchy, the 100-percent confidence loop, the Trinity rubric reference, and the output contract. Trigger when the agent first attaches to the repo, when a new turn begins after a context reset, or when the architect or user explicitly invokes "system_directive" or "load governance." Required prerequisite for setup_intake.

0
installs
▌
Workstream Creator · 0sxd
Use this skill to spawn a new workstream MD file when the user introduces a research thread, a build target, or a project area that needs its own scoped scaffold. Trigger when the user says "new workstream", "spin up a separate effort for X", "I want to track X separately", "create a workstream for Y", or when the setup_intake skill detects that the current goal does not fit in any existing workstream. Generates a workstream MD that the user fills with their research bundle, then routes to the other skills.

0
installs
▌
Edit Mode · 0sxd
Governance unlock/lock cycle. Use /sagex:edit-mode to unlock rules, skills, hooks, approved sources for modification. Use /sagex:lock to re-engage. Only Sage can invoke.

0
installs
▌
Audit Loop · 0sxd
Use when code has been written or modified. Triggers the 3-agent audit cycle: ZCR (zero-context review) → Semgrep (static analysis) → Opus (cross-reference verdict). Code cannot be committed without passing all three.

0
installs
▌
Brain Harness · 0sxd
Full session bootup sequence. Triggered when user asks about accessing memory, putting on the harness, checking the brain, or any similar phrase indicating they want full session initialization. Reads all memory files across all active projects and reports session state before asking what we're working on today.

0
installs
▌
Research Gate · 0sxd
Use when writing or editing .py files. Enforces approved-source research before code writes. Trigger: any .py file creation or modification. This skill reads config/approved_sources.json and validates that research.md contains URLs from approved domains that actually resolve. Blocks writes without valid research.

0
installs
▌
Coding Harness · 0sxd
Dispatch a Sonnet coding agent, run 5-gate audit pipeline, present results. Invoke when Opus has a coding task ready for execution. This is the local coding harness — the primary agent species for this workspace.

0
installs
▌
Worktree Isolation · 0sxd
Create and manage git worktrees for parallel coding agent isolation. Invoke when two independent coding tasks need to run simultaneously (max 2).

0
installs
▌
Notebooklm Research · 0sxd
Repeatable pattern for conducting NotebookLM research via Gemini handoff. Use when research needs to go through a NotebookLM corridor.

0
installs
▌
Corpus Inventory · 0sxd
Teaches agents how to read, query, and contribute to the corpus inventory wiki. Provides the classification taxonomy, wiki page contracts, and update protocols for maintaining the multi-agent knowledge index.

0
installs
▌
Youtube Transcript · 0sxd bundle
Fetch YouTube video transcripts (auto-generated or manual subtitles) without API keys, using the open-source youtube-transcript-api Python library, with yt-dlp as a fallback for blocked videos. Use when the user asks for a transcript / captions / subtitles of a YouTube video, or when summarizing / quoting a video.

0
installs

19 published skills