Profile My skills Saved Collections Edit profile Submit a skill

← all publishers

claudialnathan

@claudialnathan source repo

11 published skills

  1. ▌
    Openspec Explore · claudialnathan
    Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.
    0
    installs
  2. ▌
    Openspec Propose · claudialnathan
    Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.
    0
    installs
  3. ▌
    Openspec Apply Change · claudialnathan
    Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.
    0
    installs
  4. ▌
    Ingest · claudialnathan bundle
    Work out where handed-over reading belongs in a coding-agent harness: which existing primitive it improves, whether it warrants a new one, or that nothing does. Use when given articles, links, papers or someone's public writing without a named destination. Material for an artifact the owner already names is an edit to that artifact, not a placement question.
    0
    installs
  5. ▌
    New Skill · claudialnathan bundle
    Build a new coding-agent harness artifact: a skill, a hook, a permission rule, a path-scoped rule, a CLAUDE.md or AGENTS.md entry, a subagent, an MCP connection, or a workflow script. Use when making a skill about a subject, turning an article or a standing preference into one, or working out which surface a behavior belongs on. Deciding to build nothing is a normal outcome.
    0
    installs
  6. ▌
    Grill Skill · claudialnathan bundle
    Test an existing coding-agent harness artifact on a real prompt, then turn the owner's reaction into a change that holds beyond that one example. Use when trying a skill on something real and iterating from feedback, especially for voice, taste and judgment work where the owner is the only judge. Needs both the artifact and a prompt.
    0
    installs
  7. ▌
    Harness Audit · claudialnathan bundle
    Audit what a project's agent sessions load before any work starts: the instruction-file chain, visible skills, MCP instruction blocks, agent rosters, hooks and settings at every scope. Quantify what those cost per session in tokens, check whether the setup's claims about itself hold, and check whether the surfaces contradict each other once assembled. Use to review or slim a project's harness, after major harness changes, or when sessions run fine but take contradictory steering from turn one. A failing command, a startup warning, or a config that will not load is a debugging job rather than this one, and auditing the guidance inside a single artifact is a third job. Read-only, and nothing is applied without the owner's go-ahead.
    0
    installs
  8. ▌
    Improve Skill · claudialnathan bundle
    Change an existing coding-agent harness artifact. Use when a skill misses, over-fires, never triggers, or steers badly, when an article or doc should be worked into one that already exists, or when the artifact may no longer be needed at all. Deleting it is a normal outcome.
    0
    installs
  9. ▌
    Model Bump · claudialnathan
    Refresh MODELS.md against current model releases, pricing, availability, and task-specific benchmark results.
    0
    installs
  10. ▌
    State Bump · claudialnathan
    Refresh STATE.md against the live Claude Code changelog. Edits the "Last updated:" / "Version:" header on line 3, then adds, edits, or retires body entries so the snapshot matches what's at code.claude.com/docs/en/changelog. Preserves the document's existing voice (terse, factual, "what changes how you build, configure, and ship"), its section structure, and the dateline convention, which means header fields only and no inline "last reviewed against vX.Y.Z" reintroduced elsewhere. On every substantive bump, also re-pin and amend HACKS.md, check AGENTS.md for staleness, and sweep the kitchen's skills/ artifacts against the new state (Steps 5–7).
    0
    installs
  11. ▌
    Ship Agent Skills · claudialnathan
    Commit and push complete work in the agent-kitchen repo with its commit discipline, then update this machine's live Claude, Cursor, and Codex skill/plugin copies. Use for requests to commit, push, or ship changes in this repo. Invocation is consent to push and run the documented local sync and plugin update commands.
    0
    installs