agentic-repos

Record an improvement, learning, fix, or refinement to the Agentic Repos framework. Invoke when a new pattern emerges, a bug is found in a skill/agent/rule, or a workflow refinement is identified during work. Handles version bumping, CHANGELOG updates, and cross-file consistency. Also user-invocable via "ar-add-improvement" or "update framework".

Install or update context-processing skills (discussion, email, meeting, specs, state) into a Docs Project. Separate from ar-taskflow - does not touch ar-taskflow skills or config.

Adopt Agentic Repos in a project, writing its per-repo config + rules layer by extracting the codebase's real conventions. Say "ar-install" or "make this repo agent-ready".

Install or refresh the global Agentic Repos layer on this machine (ar-* skills, scripts, session hook, default-branch guard). Say "ar-install-tools" or "setup tools".

Reviews a PULL REQUEST of the Agentic Repos framework SOURCE repo itself for project-based noise. Its input is a framework PR, on trigger it FIRST asks which agentic-repos PR to review (framework PRs only; it refuses PRs from any other repo). Proves the PR introduces ZERO project-based noise into the project-agnostic framework, installed artifacts (rules/, skills/, agents/, templates/, setup.md) AND docs/. Catches a source project's identity (real names, packages, developer paths, tickets, people, dated incident notes), whole files bound to one project (a <project>-feedback.md mining doc, anything named after or dominated by a single product line, these belong in the source project, not the framework), stack idioms leaking into stack-agnostic skill/agent bodies, ungenericized business-domain examples, rationale/version-history bloat, mis-tiered rules, and version-file drift. Use whenever you want to review/self-review an agentic-repos PR for "does this leak or bind a project into the generic framework," befor

Refresh an adopted project's per-repo config + rules layer to the current framework version. Say "ar-upgrade" or "refresh the project rules".

Assess how agent-ready a repository is and raise it. Reads the actual codebase, extracts its real conventions into rule files, wires the config seam, and reports a readiness scorecard. Say "ar-agent-ready", "make this repo agent-ready", or "extract the rules for this project".

Audit what the code ASSUMES against what production ACTUALLY does, using Datadog traces, and turn mismatches into fix tickets. Say "ar-api-dd-compare" or "trace audit". On-demand only (Datadog MCP is interactively authenticated).

Scan a window of Datadog traffic (default last day, production), sample traces for APIs over the latency/error thresholds, cross-reference this repo's code, and maintain per-API optimization report files under the workspace Performance_Findings directory, then commit them. On-demand only. Say "ar-dd-api-performance", "api performance scan", "which APIs are slow", or "update the performance reports".

Review any GitHub PR from anywhere on the machine, no project checkout needed. Clones the repo, sets up a review worktree, does a best-effort test run, reviews changed code for logical bugs and missing-test gaps against the project's rules, dedups against existing comments, and posts an inline review with fixes. Say "ar-global-pr-reviewer", "review pr globally", or "global review <PR-URL>".

Initialize issue-tracker integration for Claude Code. For GitHub Issues (the default) this verifies the gh CLI is authenticated, no MCP needed. For Jira (Atlassian) or Linear it sets up the Model Context Protocol (MCP) server. Say "ar-init-mcps" or "setup mcps".

Initialize skills configuration for this project. Use when setting up for first time or when skill.config is missing. Say "ar-init-skills" or "setup skills".

Audit and optimize a project's AI rule files. Detects redundancies, conflicts, staleness, example/rationale bloat, and fragmentation. Produces detailed reports and interactive cleanup recommendations. Say "ar-optimizer" or "optimize AI rules" to run.

Record a framework improvement suggestion from any target project. Saves structured improvement files to the workspace's shared improvements/ directory for ar-add-improvement to consume later. Say "ar-record-improvement" or "record improvement".

Aggregate and fix PR feedback from SonarQube, CodeRabbit, and human reviewers in priority order, then reply to and resolve the review threads. Standalone skill - run after PR feedback arrives. Say "ar-taskflow-fix-comments" or "fix comments".

Plan a large feature with issue-tracker integration. Creates a planning ticket immediately, guides architecture discussion, writes spec, breaks into raw prompts, creates a Story ticket and story branch on approval. Say "ar-taskflow-planner" or "plan feature".

Use when Claude forgets context or hallucinates during the same session. Re-reads task-flow skill and task files to get back on track. Say "ar-taskflow-remember" or "remember".

Resume an ongoing task from where you left off. Use when user says "ar-taskflow-resume" or "resume task". Asks for task folder path and continues from the last phase.

Reviews code changes in ar-taskflow workflow against coding rules. Integrates with task folders, tracks iterations. Use after code implementation. Say "ar-taskflow-review" or "review".

Structured development workflow from raw prompt to PR. Use when user says "ar-taskflow" or provides a task folder path. Follows Raw Prompt - Understand - Plan - Code - Document flow.

Create one well-formed, PR-sized ticket from your current work, using the project's configured issue tracker. Say "ar-ticket-creator", "create ticket", or "spin off a ticket". For large multi-PR features use ar-taskflow-planner instead.

Reconcile workspace state against reality. Catches dangling Phase-4 tasks not archived, mismatched branches, stale execution-summary state, missing TasksSummary rows, PR-merge gaps, and docs-repo drift. Renamed from ar-taskflow-inspector in v6.6.0. Say "ar-taskflow-progress-fixer" or "reconcile tasks".

Generate a defensible weekly status report from task tracking artifacts. Audience-configurable (CEO/CTO, Engineering Lead, PM, etc.). Reads TasksSummary, WeeklySummaries, ProjectState, and per-task folders to produce evidence-grounded prose. Say "ar-weekly-report" or "weekly report".