johnend/dots

johnend/dots

Skill Claude CodeCodex

Brief description of what this skill does and when the agent should use it. Be specific; include trigger terms so the agent knows when to apply.

johnend/dots

Skill Codex

Review the latest Claude Code documentation, compare it against the current Claude Code configuration, and recommend specific improvements based on new or underused features. Run periodically (e.g. monthly) to catch new capabilities that could improve the workflow. TRIGGER when: user asks to audit, review, or check…

johnend/dots

Skill Codex

Draft or update high-quality technical documentation in the Obsidian vault, with clear structure, Obsidian-ready frontmatter, tags, rationale, and examples. TRIGGER when: user asks to document something, write or update docs/notes, add a page to the Obsidian vault / knowledge base, "chronicle this", or "write this…

johnend/dots

Skill Claude CodeCodex

Create a pull request OR update the description of an existing PR for the current branch. Handles repo PR templates, JIRA ticket extraction (including title prefix), a prose summary followed by implementation bullets with rationale, testing evidence, and screenshot prompts for UI/frontend changes. On FanDuel repos…

johnend/dots

Skill Codex

Draft a tech/Jira ticket from conversation context or user-provided details, get explicit approval, then create it — as a real Jira issue (FanDuel repos) or an Obsidian note (personal projects). TRIGGER when: user asks to write/create/draft a ticket, "file a bug", "make a Jira ticket", "write this up as a…

johnend/dots

Skill Codex

Create or update an agent skill so a durable behavior change lands in the right SKILL.md, with proposal-first approval. TRIGGER when: user asks to create/update/change a skill, "turn this into a skill", "make a skill for this", or gives feedback about how you should behave in future sessions. DO NOT TRIGGER for…

johnend/dots

Skill Claude CodeCodex

Fetch review and discussion comments on the current branch's pull request, classify human-reviewer feedback against tracked patterns, and turn it into a concise action list. TRIGGER when: user asks to get/fetch/summarize PR comments, "what did reviewers say", "any feedback on my PR", check review comments, or address…

johnend/dots

Skill Claude CodeCodex

Behavioral checklist to avoid common LLM coding mistakes: resist overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria. TRIGGER when: about to start a non-trivial implementation, refactor, or debugging task — read before writing code; also when the user asks you to…

johnend/dots

Skill Codex

Hypothesis-driven root-cause investigation for bugs with unclear causes: pin scope, generate competing hypotheses, try to falsify each with telemetry and code, keep a ledger, and record findings in the Obsidian work item as you go. Docs-only until the diagnosis is confirmed. TRIGGER when: user asks to investigate…

johnend/dots

Skill Claude CodeCodex

Add or remove CLI/tool patterns in Claude Code's sandbox permissions allowlist and denylist in settings.json. TRIGGER when: a Bash command is blocked by the sandbox, or the user asks to allow/deny a command, "add X to the allowlist", "stop asking permission for Y", or "sandbox blocked this". DO NOT TRIGGER for…

johnend/dots

Skill Codex

Tidy a mostly-finished patch before review or commit: remove AI-style slop, check test value, validate key paths, and surface remaining risks. Edits code. TRIGGER when: user asks to clean up / tidy / polish a patch or diff before committing, "remove the slop", "get this ready to commit", or "clean this up before I…

johnend/dots

Skill Codex

Review the current session for lessons (corrections, re-explanations, repeated failures, tool workarounds) and route each one to the right durable home: Applied Learning in /.ai/AGENTS.md, project memory, a skill, or a hook. TRIGGER when: user says "retro", "what did we learn", "capture the lessons", "record that"…

johnend/dots

Skill Codex

Review the user's own unmerged work for bugs, regressions, risk, and missing tests with severity-first output and concrete file references. Handles two scopes: uncommitted/staged changes (pre-commit) and the full branch diff against its base (pre-PR self-review). Read-only — produces findings, does not edit code.…

johnend/dots

Skill Codex

Review a colleague's pull request and output a structured review with actionable findings — no auto-commenting on GitHub, no plan mode, no implementation. TRIGGER when: user asks to review a PR, review a pull request, check a PR, look at a PR, give feedback on a PR, first-pass review, or references a PR number/URL to…

johnend/dots

Skill Claude CodeCodex

Produce a rich end-of-session Obsidian note — what was done, why, patterns implemented, and codebase knowledge gained — then hand off to chronicle-docs to format and store it. TRIGGER when: user asks to wrap up the session, write session notes, "document what we did", or "capture this session" before finishing. DO NOT…

johnend/dots

Skill Codex

Step-by-step tutorial guidance for a TypeScript-comfortable engineer learning something new: the AI explains, the user learns. Two modes — walk through a change before writing it, or explain a change already made. Never edits files or writes the implementation. TRIGGER when: user asks to be taught, walked through, or…

johnend/dots

Skill Codex

Support-rota triage: take a reported failure (Slack message, staging/QA bug, error screenshot), trace it to the failing component, identify the owning team from evidence, and draft a short Slack hand-off. TRIGGER when: user is on support/rota duty, pastes a Slack report or error and asks what's going on, "who owns…