dev10x-qa
claudev0.115.0 — 6 skills for self-QA walkthroughs: Playwright capture, narration, evidence publishing, YouTube upload, and review packs.
6 skills
v0.115.0 — 97 skills for structured commits, PR lifecycle, branch hygiene, ticket management, task tracking, code review, and scoping in Claude Code.
Plugin dev10x from marketplace Dev10x-Guru in Dev10x-Guru/Dev10x-Claude. Commands read from the repo's manifest — 5 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use and it starts local MCP servers. The exact commands are listed below; read them before installing.
dev10x:adr-evaluateOrchestrate adversarial evaluation of architectural options using domain-specific architect agents and produce ranked ADR drafts. TRIGGER when: evaluating design decisions with multiple competing approaches, running architecture trade-off analysis, or needing structured multi-perspective evaluation.
dev10x:adrCreate Architecture Decision Records (ADRs) following project conventions. Extends the base scope skill with ADR-specific format, numbering, diagram generation, and decision documentation. TRIGGER when: documenting significant architectural decisions, design trade-offs, or technology choices that af
dev10x:afkWalk-away mode — compose the walk-away gate policy so long-running sessions do not stall on re-strategy or confirmation gates. Writes gate_preset: adaptive and gate_overlays: [afk] to the global ~/.config/Dev10x/friction.yaml via `dev10x session set-friction` (ADR-0018); the resolve_gate resolver re
dev10x:artifactUse when someone asks for an Artifact page — a report, analysis, status page, rollout review or decision page — so it lands on the first try (full width, sectioned aside, answer first, diagrams that render in both themes, evidence) instead of after rounds of layout and diagram corrections. TRIGGER w
dev10x:askSurface what is still open in a session and route it where it survives. Reformulate plain-text decision questions into AskUserQuestion widgets, sweep the session for open loops and sync them to the task list, or output a reinforcement nudge. TRIGGER when: supervisor sees a plain-text decision questi
dev10x:audit-fileFile skill-audit findings as a GitHub issue at the plugin repo that owns the offending skill. Invoked by skill-audit Phase 7 with the target repo confirmed at its destination gate. TRIGGER when: skill-audit Phase 7 completes and user opts to file upstream. DO NOT TRIGGER when: no audit findings exis
dev10x:busLet this session exchange messages with other Claude sessions on the same machine through a local Maildir bus — other Claude Code sessions, and Claude Desktop / claude.ai chats (text or voice) whose Cowork agent writes the files via Desktop Commander. Opens a mailbox under /tmp/claude-bus/<name>/, a
dev10x:ci-reviewUse when a change to a Claude CI review prompt, args file, reviewer agent or CLI pin needs checking before it merges — so you read what the CI code review and PR hygiene check would post on a real PR, with nothing reaching GitHub unless you choose, after reading the report, to share it as one marked
dev10x:context-auditAudit context window utilization — CLAUDE.md files, rules, agent specs, references, and memories. Reports budget compliance, flags oversized files, and suggests pruning actions. TRIGGER when: user wants to optimize context usage, after adding many rules or memories, or when sessions feel sluggish fr
dev10x:dddRun or continue a DDD Event Storming workshop to explore, model, and stress-test domain architecture. TRIGGER when: user mentions DDD, domain modeling, event storming, bounded contexts, domain events, aggregates, "workshop", "domain session", "stress test the architecture", "apply archetype", "scope
dev10x:diag-approvalUse right after manually approving a permission prompt — the way /dev10x:diag-friction is used after a prompt is refused — so the approval you just gave becomes a pre-approval everywhere, or you learn why a skill, a rewrite or a directory setting fixes it better, without retyping the command. TRIGGE
dev10x:diag-frictionDiagnose permission friction and end in a fix: a rewrite, a wrapper, a backfill, or a pre-approval written everywhere through dev10x:permission-grant, so the same prompt does not come back next session. Guides the agent toward pre-approved commands, simplifies complex command chains, and (when no sa
dev10x:diagram-bpmnUse when someone asks to model a business process, workflow, approval chain, or hand-off between roles as BPMN — a file that opens in Camunda Modeler or bpmn.io, plus a rendered image for a doc or PR. TRIGGER when: drawing who does what, in which order, with decisions and loops, across roles or lane
dev10x:diagram-c4Use when someone asks for an architecture diagram, a system context or container view, "what talks to what", a component breakdown of one service, or a deployment picture — for an ADR, README, design doc, or onboarding page. TRIGGER when: drawing system structure at C4 level 1-3, dynamic, or deploym
dev10x:diagram-sequenceUse when someone asks for a sequence diagram, a call flow, a request/response timeline, or "who calls whom" between services, agents, or APIs — for a PR description, issue, ADR, or design doc. TRIGGER when: drawing interactions between participants over time. DO NOT TRIGGER when: drawing system stru
dev10x:estimateEstimate how long a ticket or project will take when AI agents do the implementation and humans still gate review, testing and release — so a lead plans on numbers that match how the work actually ships, instead of one engineer's human-pace guess. TRIGGER when: sizing a ticket or project, filling an
dev10x:fanoutProcess multiple independent work items as a worktree-isolated swarm — each item runs the full dev10x:work-on lifecycle inside its own background Agent with isolation="worktree" (under Herdr, its own Herdr workspace and agent). Honors dependencies, minimizes conflicts, auto-advances by default. TRIG
dev10x:foremanUnattended overnight delivery of a milestone/bundle queue — a two-tier harness (watchdog main session + cheap foreman overseer + work-on crew) that pre-flights permissions while the supervisor is still present, loops the queue, defers questionable scope to the end, survives quota-block rollovers, an
dev10x:friction-setupGuide the supervisor through the one question the model needs — does the supervisor read the PR before the next step is allowed? — via a blocking AskUserQuestion gated on `pinned: false`, plus the existing solo-vs-team overlay / per-gate-deviation / skippable-step walk, then persist the choices to t
dev10x:gchat-review-requestPost a Google Chat review request for a PR using per-repo config (space, mentions). Mirrors dev10x:slack-review-request. Invoked standalone or delegated to by dev10x:request-review. TRIGGER when: a PR needs a Google Chat review notification. DO NOT TRIGGER when: Google Chat is not configured, or pos
dev10x:gchatSend notifications to a Google Chat space via a private Chat bot (service-account app auth). Mirrors dev10x:slack. TRIGGER when: sending a message to a Google Chat space. DO NOT TRIGGER when: posting a review request (use dev10x:gchat-review-request), or sending to Slack (use dev10x:slack). - Bash(u
dev10x:gh-contextDetect PR context (number, repo, URL, branch) from a URL, PR number, or current branch — so skills like dev10x:gh-pr-monitor always get the correct target PR even in multi-worktree setups. TRIGGER when: resolving PR context from URL, number, or branch for downstream skills. DO NOT TRIGGER when: PR c
dev10x:gh-pr-bookmarkPost a rich session bookmark comment on a PR — captures session ID, review thread status, current state, and next steps so the next session can pick up where this one left off. TRIGGER when: pausing or ending session with an open PR that has in-progress work. DO NOT TRIGGER when: no PR exists, or se
dev10x:gh-pr-createCreate a GitHub pull request for the current branch with issue tracker integration (GitHub Issues, Linear, or JIRA). Sources or generates a JTBD Job Story for the PR description, extracts ticket info from branch name, pushes the branch, creates a draft PR with Job Story, commit list, and issue track
dev10x:gh-pr-doctorUse when merged PRs may have unresolved review threads accumulating as silent tech debt — so unaddressed feedback gets surfaced, grouped by theme, and tracked as issues. TRIGGER when: reviewing merged PRs for accumulated tech debt, or post-merge as part of maintenance workflow DO NOT TRIGGER when: r
dev10x:gh-pr-fixupImplement a fix for a validated PR review comment, create a fixup! commit, push, and reply with the commit reference. Creates ONE fixup commit per comment. TRIGGER when: PR review comment has been validated as needing a code fix. DO NOT TRIGGER when: comment is invalid/question (use dev10x:gh-pr-tri
dev10x:gh-pr-mergeValidate all pre-merge conditions and execute PR merge. Checks unresolved threads, CI status, draft state, mergeability, working copy, fixup commits, Fixes-linked scope delivery, and review approval before merging. TRIGGER when: PR is ready to merge and needs pre-merge validation. DO NOT TRIGGER whe
dev10x:gh-pr-monitorLaunch a background agent to monitor PR CI checks and review comments, automatically address issues with fixup commits, and notify team when ready. Use after creating a PR to automate the entire review cycle. TRIGGER when: PR has been created and needs CI/review monitoring. DO NOT TRIGGER when: PR d
dev10x:gh-pr-request-reviewRequest review on a GitHub PR from teams or users. TRIGGER when: PR is ready for review and needs reviewer assignment. DO NOT TRIGGER when: PR is still draft or WIP, or review was already requested. - mcp__plugin_dev10x_cli__request_review - mcp__plugin_dev10x_cli__resolve_gate - mcp__plugin_dev10x_
dev10x:gh-pr-respondValidate and respond to PR review comments. Handles single comment (with follow-up offer) or batch mode for all unaddressed comments on a PR/review. Orchestrates dev10x:gh-pr-triage and dev10x:gh-pr-fixup. TRIGGER when: PR has review comments that need responses or fixes. DO NOT TRIGGER when: no rev
dev10x:gh-pr-reviewReview a GitHub pull request and post findings with inline comments. Fetches PR diff, reads changed files, checks for interface impact, applies project review guidelines, and posts a review to GitHub. Supports Draft (PENDING) and submitted review modes via Step 8a gate. Supports courtesy-fixup dispo
dev10x:gh-pr-triageValidate a PR review comment against the codebase. If invalid, reply with evidence. Never auto-resolves threads — resolution requires explicit user confirmation. Returns a verdict (VALID, YAGNI, INVALID, QUESTION, OUT_OF_SCOPE) so the caller knows whether to fix, remove out-of-scope code, reply, or
dev10x:gh-review-setupGuided, discovery-driven setup of Claude code-review GitHub Actions on any repo. Detects the stack, lets you pick independently-skippable review modules (code review, PR hygiene, ADR distillation, lessons-learned, second-opinion), surfaces pros/cons for each strategy decision, then scaffolds general
dev10x:git-alias-setupSet up git aliases that reduce permission friction by wrapping $(git merge-base ...) subshells into stable command prefixes. TRIGGER when: SessionStart reports missing aliases, or user hits permission friction from subshell-based git commands. DO NOT TRIGGER when: aliases already configured (check g
dev10x:git-branch-pruneClassify and prune stale local branches with merge verification, designed for rebase-merge repos where `git branch --merged` misses most merged branches. TRIGGER when: local repo has accumulated stale branches after fanout sessions, worktree cleanup, or manual sweeps. DO NOT TRIGGER when: working on
dev10x:git-commit-splitSplit monolithic git commits into atomic, cohesive commits following Clean Architecture principles. Uses interactive rebase to separate changes by feature dependency order (utilities → data → DTOs → refactoring → features → API), ensuring each commit is self-contained, passes tests, and maintains pr
dev10x:git-commitCreate a properly formatted git commit following project conventions (gitmoji, ticket reference, 72 char limit). Extracts ticket ID from branch name, prompts for description and solution points, stages changes, and creates the commit. TRIGGER when: creating a git commit with proper formatting. DO NO
dev10x:git-fixupCreate a fixup! commit for a PR review comment or standalone improvement. Enforces one fixup per comment thread when linked to a review. TRIGGER when: creating a fixup! commit for a review finding or standalone improvement. DO NOT TRIGGER when: creating a regular commit (use dev10x:git-commit), or i
dev10x:git-groomRestructure, polish, and clean up git commit history in the current branch before merging. Creates atomic, well-organized commits that tell a clear story. TRIGGER when: branch is ready for merge and commit history needs cleanup (squash fixups, reorder, reword). DO NOT TRIGGER when: branch has clean
dev10x:git-promoteConvert existing git commits into issue tracker tickets with proper branch management and commit message formatting. Automates creating the ticket, branching, cherry-picking, and updating the commit message. TRIGGER when: a commit needs to be retroactively tracked with a ticket, typically for tech d
dev10x:git-worktreeCreate git worktrees for clean workspace isolation. Offers two modes: native EnterWorktree (switches CWD in current session) or external worktree (IDE-isolated, requires restarting claude in new dir). Inside Herdr (HERDR_ENV=1) a third mode is chosen automatically: Herdr creates the worktree in its
dev10x:gitUse before running git push or git rebase — so force-pushes to protected branches are blocked and non-interactive rebases run unattended without manual editor approval prompts. TRIGGER when: running git push or git rebase operations. DO NOT TRIGGER when: other git operations (commit, status, log, di
dev10x:gogReach Google Workspace from the command line through the `gog` CLI — Drive, Google Chat, YouTube, Gmail, Calendar — including the OAuth setup that every one of those areas shares and that fails in the same four ways for everybody. TRIGGER when: reading or writing Google Workspace data from a session
dev10x:ide-normalizeNormalize PyCharm .idea/ configuration after creating or copying a new git worktree. Fixes stale module-name references, disables ADD_CONTENT_ROOTS / ADD_SOURCE_ROOTS defaults that conflict with editable installs, backfills the Django facet settingsModule, and detects the PyCharm uv-SDK FLAVOR_DATA
dev10x:investigateRoot-cause a bug report, question, or unexpected behaviour from a Slack thread — posts a technical reply with GitHub links and creates a Linear ticket when warranted. TRIGGER when: given a Slack URL pointing to a bug report, question, or unexpected behaviour that needs investigation. DO NOT TRIGGER
dev10x:jiraUse when querying, linking, or fetching JIRA issues — so credentials, hook-safe curl patterns, and hierarchy gotchas are always at hand
dev10x:jtbdPure JTBD story drafting skill. Gathers context from issue tracker tickets, parent tickets, and PR diffs to craft a situation-driven Job Story with explicit actor and beneficiary. Returns the draft string with no side effects. TRIGGER when: drafting a JTBD Job Story for a ticket, PR, or release. DO
dev10x:linearLinear issue tracker operations via MCP tools. Get, create, update issues and list comments. Documentation-only skill — no scripts. TRIGGER when: performing CRUD operations on Linear issues, comments, or statuses. DO NOT TRIGGER when: using GitHub Issues or JIRA (use dev10x:jira), or creating ticket
dev10x:memory-maintenanceAudit project and global memory files for stale patterns, version-pinned paths, contradictory instructions, and anti-patterns that cause skill bypass or permission friction. Reports findings and offers to clean up. TRIGGER when: user wants to audit memory health, after a plugin upgrade, or when agen
dev10x:onboardingGuided discovery of Dev10x capabilities for new users. Interactive tour through skill families, git setup, PR pipeline, session management, and customization — adapted to experience level. TRIGGER when: user is new to Dev10x, asks what it can do, or wants a walkthrough of available capabilities. DO
dev10x:overwatchRun the localhost Overwatch page that lists every local Claude Code session with its state, queues the questions sessions wait on, and lets the supervisor answer them in one place — or run the relay session that carries those answers back into the right session with SendMessage. TRIGGER when: the us
dev10x:park-discoverGather deferred items across all sources — so nothing is missed when starting a session or picking up where you left off. TRIGGER when: starting a session, picking up prior work, or checking for deferred items. DO NOT TRIGGER when: mid-session active work with no need to check deferred items. - Read
dev10x:park-remindSchedule a Slack reminder — so deferred items appear when you are clearing messages, not buried in a file you might not open. TRIGGER when: deferring work that should resurface via Slack notification later. DO NOT TRIGGER when: deferring to code or project storage (use dev10x:park-todo), or routing
dev10x:park-todoDefer work to code or session-level storage — so items resurface when editing nearby code or starting a new session in the same project, instead of being forgotten. TRIGGER when: deferring work to code comments or the project task index. DO NOT TRIGGER when: deferring to Slack (use dev10x:park-remin
dev10x:parkSmart deferral router — saves tasks for later to the right place (PR, ticket, code, Slack, or the project task index) so they are actually rediscovered instead of being forgotten. TRIGGER when: a task should be saved for later instead of done now. DO NOT TRIGGER when: task should be done now, or spe
dev10x:permission-audit-applyUse when a permission-auditor report has landed and its proposed fixes need applying — so each accepted fix becomes one approved edit and nothing outside the approved scope rides along, instead of the user hand-editing settings files finding by finding. TRIGGER when: dev10x:plugin-maintenance step 1
dev10x:permission-grantUse when a useful command keeps prompting and should be pre-approved everywhere at once — the global settings, every project and every worktree — so one decision ends the prompt in every session instead of one file at a time, and a command that a skill, a rewrite or a directory setting would fix bet
dev10x:permission-investigatorMaterialize a fixture, mutate settings with candidate rule shapes, and dispatch subagents to record which shapes the permission engine auto-approves vs prompts. Aggregates results into a markdown matrix and computes a delta against the rule shapes shipped by dev10x:plugin-maintenance. TRIGGER when:
plan-syncplaybook-maintenanceplaybookplugin-doctorplugin-maintenanceproject-auditproject-scopepy-mockitopy-test-flakypy-testpy-uvqa-scopereflectrelease-notesrequest-reviewreview-fixreviewscopesession-config-seedsession-taskssession-wrap-upskill-audit-queueskill-auditskill-createskill-indexskill-pruneslack-review-requestslack-setupslackspec-syncspec-updateticket-branchticket-createticket-jtbdticket-scopeupgrade-cleanupverify-acc-dodvoice-channelwork-onadr-reviewerSynthesize arguments from architect agents into ranked ADR options. Fact-checks claims against the actual codebase. Used by dev10x:adr-evaluate skill during the synthesis phase. Triggers: invoked by dev10x:adr-evaluate skill only
architect-apiEvaluate API design decisions — REST, GraphQL, gRPC patterns, schema design, real-time strategies, and API layer compliance. Used by the adr-evaluate skill for API-related ADRs. Triggers: invoked by adr-evaluate skill for API design ADRs
architect-dbEvaluate database architecture decisions — schema design, indexing, multi-tenant strategies, JSONB patterns, migration safety, and query optimization. Used by adr-evaluate for data ADRs. Triggers: invoked by adr-evaluate skill for database architecture ADRs
architect-domainEvaluate domain modeling decisions using DDD principles, Software Archetypes, and Clean Architecture patterns. Used by adr-evaluate for domain modeling ADRs. Triggers: invoked by adr-evaluate skill for domain modeling ADRs
architect-frontendEvaluate frontend architecture decisions — component patterns, state management, SSR strategies, testing approaches, and design system patterns. Used by adr-evaluate for frontend ADRs. Triggers: invoked by adr-evaluate skill for frontend architecture ADRs
architect-infraEvaluate infrastructure decisions — deployment strategies, CI/CD pipeline design, Docker patterns, observability, and environment management. Used by adr-evaluate for infrastructure ADRs. Triggers: invoked by adr-evaluate skill for infrastructure ADRs
architecture-advisorUse this agent when you need to evaluate software architecture, identify design issues, or plan new component designs. Covers Clean Architecture, DDD, SOLID principles, and enterprise patterns. Triggers: "review the architecture", "design a new component", "evaluate this design", "identify tech debt
code-reviewerUse this agent to review code changes in the current Git branch according to project standards and best practices. Reviews only changed files — not the entire codebase. Triggers: "review my code", "check this before PR", "code review" Do NOT use for: reviewing entire codebases, writing code, fixing
infrastructure-investigatorUse this agent to investigate Kubernetes infrastructure, troubleshoot deployment issues, analyze observability stacks, check resource states across environments, or find discrepancies between git configs and actual cluster state. Triggers: "check k8s", "why is this pod failing", "compare staging and
issue-investigatorUse this agent to investigate bugs, errors, performance issues, or unexpected behavior in any project. Performs deep-dive analysis including database forensics, code path tracing, timeline reconstruction, and root cause hypothesis generation. Triggers: "investigate this bug", "why is this failing",
permission-auditorUse this agent when you need to audit Claude Code permission settings for security gaps, overly broad allow rules, missing deny rules, unregistered hooks, privilege escalation paths, and script-path leaks in instruction files. This agent performs a comprehensive 7-phase analysis of settings.local.js
pytest-test-writerUse this agent to write, review, or improve Python unit tests using pytest. Covers fixture design, mockito stubs, parametrization, and comprehensive coverage patterns. Triggers: "write tests for", "review my tests", "help with fixtures", "test coverage for
pytest-testerUse this agent to run tests and verify coverage before committing, when creating a PR, or after writing new code. Ensures test suite passes and provides concise summaries. Triggers: "run tests", "check coverage", "verify tests pass", "ready to commit
reviewer-celeryReview Celery task definitions for naming, periodic registration, discovery, and data migration safety. Triggers: files matching **/tasks.py, **/celery.py
reviewer-e2eReview Playwright page objects and behave feature files for step ambiguity, locator fragility, and fixture alignment. Triggers: files matching **/features/**/*.feature, **/features/steps/*.py, **/environment.py, **/e2e/**/*.py
reviewer-frontendReview frontend files — components, routes, i18n, auth patterns, accessibility, and SSR safety. Triggers: files matching **/*.svelte, **/*.astro, **/*.tsx, **/*.jsx
reviewer-graphqlReview GraphQL schema changes for type safety, backward compatibility, authorization, and Clean Architecture compliance. Triggers: files matching **/api/queries.py, **/api/mutations.py, **/schema.py
reviewer-migrationReview database migration files for data loss, backward compatibility, lock contention, and tenant scoping. Triggers: files matching **/migrations/*.py
reviewer-signalsReview signal handlers, event receivers, and repository lookup semantics for correctness, naming, and exception safety. Triggers: files matching **/signals.py, **/handlers.py, **/receivers.py
reviewer-test-flakyDetect test flakiness risks: random ID collisions, low-entropy providers, constraint races, and missing branch coverage. Triggers: files matching **/tests/**/*.py
reviewer-test-patternsReview test files for pattern compliance, coverage gaps, fixture DRY, mockito teardown safety, and parametrization best practices. Triggers: files matching **/tests/**/*.py
spec-reviewerVerify a code change matches its declared specification BEFORE domain reviewers spend tokens on quality review. Compares the diff against the linked Linear/JIRA/GitHub ticket's acceptance criteria or the PR's Job Story. Triggers: invoked by dev10x:gh-pr-review and dev10x:review as Phase 0 spec gate.
cliruns: ${CLAUDE_PLUGIN_ROOT}/servers/cli_server.py
dbruns: ${CLAUDE_PLUGIN_ROOT}/servers/db_server.py
$CLAUDE_PLUGIN_ROOT/hooks/scripts/audit-wrap session-start $CLAUDE_PLUGIN_ROOT/hooks/scripts/session-start.py
$CLAUDE_PLUGIN_ROOT/hooks/scripts/audit-wrap skill-pre $CLAUDE_PLUGIN_ROOT/hooks/scripts/skill-pre.py
$CLAUDE_PLUGIN_ROOT/hooks/scripts/audit-wrap validate-bash $CLAUDE_PLUGIN_ROOT/hooks/scripts/validate-bash-command.py
$CLAUDE_PLUGIN_ROOT/hooks/scripts/audit-wrap validate-bash $CLAUDE_PLUGIN_ROOT/hooks/scripts/validate-bash-command.py
$CLAUDE_PLUGIN_ROOT/hooks/scripts/audit-wrap validate-edit-write $CLAUDE_PLUGIN_ROOT/hooks/scripts/validate-edit-write.py
$CLAUDE_PLUGIN_ROOT/hooks/scripts/audit-wrap session-stop $CLAUDE_PLUGIN_ROOT/hooks/scripts/session-stop.py
$CLAUDE_PLUGIN_ROOT/hooks/scripts/audit-wrap permission-denied $CLAUDE_PLUGIN_ROOT/hooks/scripts/permission-denied.py
$CLAUDE_PLUGIN_ROOT/hooks/scripts/audit-wrap skill-post $CLAUDE_PLUGIN_ROOT/hooks/scripts/skill-post.py
v0.115.0 — 6 skills for self-QA walkthroughs: Playwright capture, narration, evidence publishing, YouTube upload, and review packs.
6 skills
v0.115.0 — 3 skills for read-only database work: query planning, a validated psql wrapper, and SQL handoff for changes a person runs.
3 skills
v0.115.0 — 2 skills for read-only infrastructure access: AWS secrets and Kubernetes inspection through aws-vault.
2 skills
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Harness-native ECC operator layer - 71 agents, 314 skills, 98 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
200 skills · 98 commands · 71 agents · 1 MCP server · hooks
Commands for git commit workflows including commit, push, and PR creation
3 commands
agentic-flow (external: maintained in ruvnet/agentic-flow, pinned to commit d19737b, plugin 3.4.0): the agentic-flow MCP server pinned to [email protected] (ReasoningBank learning, ecosystem tools prompt_injection_scan, swarm_partition_tasks, sublinear_solve, sublinear_pagerank and ecosystem_status, multi-provider routing, autopilot learning; readonly profile by default, no tool pre-approved), 2 agents, 4 skills, /agentic-flow-status, and a mod guard on agentic_flow_agent with /agentic-flow-guard
4 skills · 1 command · 2 agents · 1 MCP server
Perform AI code review on Git diffs — supports workspace changes, branch ranges, and single commits with concurrent per-file analysis, codebase search, and deep context-aware review.
2 commands
Investigate incidents, review code and execute UAT for release readiness, scan code for vulnerabilities, and run penetration tests with AWS DevOps Agent and AWS Security Agent.
13 skills · 9 commands · 1 MCP server
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.