optimiziramsi

optimiziramsi/skills

Plugin Claude Code

optimiziramsi's Claude Code toolkit — reusable skills, commands, agents, and hooks shared across projects, shipped as a single plugin.

optimiziramsi/skills

Plugin Claude Code

optimiziramsi's Claude Code toolkit as ONE plugin, organized topic-first — one folder per concern (git, commit, reporting, session, review, repo, instructions, flow, patterns, worktree, setup), each owning its own skills/commands/agents/hooks. Opinionated concerns carry env kill-switches; guards self-gate on project…

optimiziramsi/skills

Command

Commit — bare imperative single line, topic-close + pause-for-review cadence, safe staging (runs the commit skill).

optimiziramsi/skills

Hook

Runs before the agent uses a tool for Bash tool calls, executing commit-format.py via python3. From optimiziramsi/skills.

optimiziramsi/skills

Hook

Runs when the agent finishes a response, executing commit-nudge.py via python3. From optimiziramsi/skills.

optimiziramsi/skills

Skill Claude CodeCodex

How and when to make git commits. Use proactively — commit at topic close (one logical topic = one commit), then pause for the user's review before starting the next topic, UNLESS the user has opted into commit-as-you-go. Also use whenever explicitly committing or saving changes, or when a git commit is about to…

optimiziramsi/skills

Command

Run a long chat-drives-decisions / looper-executes collaborative session (runs the collab skill).

optimiziramsi/skills

Command

Process the user's .todo inbox into the agent-owned .agent/FEEDBACK.md ledger and work the items now (runs the feedback skill).

optimiziramsi/skills

Command

Prepare a .agent/grind/ incremental mission file and hand off the grind runner command (runs the grind skill).

optimiziramsi/skills

Command

Prepare .agent/loop/ background job files and hand off the loop runner command (runs the looper skill).

optimiziramsi/skills

Command

Create, switch, or close a multi-session .agent/milestone/ initiative (runs the milestone skill).

optimiziramsi/skills

Command

Create + manage single-topic .agent/plan/ implementation plans (runs the plan skill).

optimiziramsi/skills

Command

Sweep every reference to a feature being cut/deferred and propose delete-or-defer for each (runs the scope-cut skill).

optimiziramsi/skills

Command

Walk .todo / .todo-inbox and route each item to a plan, milestone, scope-cut, scheduled task, or the bin (runs the triage-todo skill).

optimiziramsi/skills

Skill Claude CodeCodex

Long collaborative sessions where the human drives architectural decisions and the AI executes via looper. Use when the user wants to review and improve a codebase across multiple topics, says "let's go through the code", "I want to address things", "long session", "keep context clean", or signals an iterative…

optimiziramsi/skills

Skill Claude CodeCodex

Process the user's .todo inbox into an agent-owned .agent/FEEDBACK.md ledger and work the items now. Use when the user says "process feedback", "feedback round", "work through my comments/notes", or mentions new items in .todo they want handled this session. Differs from triage-todo (which routes items into…

optimiziramsi/skills

Skill Claude CodeCodex

Prepare a grind mission file and hand off to the flow grind runner (.agent/bin/grind). Grind runs the SAME mission N times — each iteration a fresh Claude session that picks 1–3 new items per the mission's picking rules, commits them, and appends to a log. Use grind for large incremental backlogs the user can't…

optimiziramsi/skills

Skill Claude CodeCodex

Prepare and manage loop jobs — background tasks executed sequentially by the flow loop runner (.agent/bin/loop) using the Claude CLI. Use whenever the user mentions: loop jobs, looper, background jobs, overnight tasks, batch tasks, job queue, breaking work into pieces for automated sequential execution, or…

optimiziramsi/skills

Skill Claude CodeCodex

Create, switch, and close milestones in .agent/milestone/. Milestones are strategic initiatives that span multiple topics and sessions (e.g. "cleanup and patterns", "ship v1 iOS app"). Each milestone is a folder with its own scope, steps, progress, and notes — giving agents a tight context window per initiative. Use…

optimiziramsi/skills

Skill Claude CodeCodex

Create and manage implementation plans in .agent/plan/ files. Use this skill when: starting a new feature or refactor, the user says "plan this", "think about how to", "design the approach", "prepare a plan", or any time structured planning is needed before execution. Plans are the source of truth for a topic — they…

optimiziramsi/skills

Skill Claude CodeCodex

When a feature is being removed from the release or deferred, sweep the codebase + docs + cron + todo for every reference and propose delete-or-defer for each. Use when the user says: "cut X", "remove X from V1", "X is out of scope", "scope-cut X", "we're not building X", "drop X", or signals a feature is being…

optimiziramsi/skills

Skill Claude CodeCodex

Process items in .todo + .todo-inbox and convert each into a milestone, plan, scheduled task, scope-cut, or deletion. Use when the user says: "triage todo", "process todo", "clear todo", "address todo items", "go through .todo", "what's in todo", or signals they want the parking lot cleared. Pulls items off the list …

optimiziramsi/skills

Command

Test-first hotfix for a deployed bug — failing test, minimal fix, cherry-pick to both branches, remote ops handed to the user (runs the hotfix skill).

optimiziramsi/skills

Skill Claude CodeCodex

Test-first hotfix protocol for bugs on a deployed environment: reproduce as a failing test, diagnose, minimal fix, then land it on BOTH the mainline and the deployed/release branch via cherry-pick — with all remote git handed to the user. Use when the user reports a bug on prod/staging ("production is broken", "bug on…