Tools & Automation

CLI tools, utilities, converters, workflow automation, and productivity helpers

skills in category: 46,767
pages available: 1,949

Security grades appear on each card once the skill has been scanned. Newly imported skills may briefly show without a grade until the backfill job runs.

Open in full browser

Browse tools & automation skills

Showing 1,297–1,320 of 46,767 skills

Copilot Sdk Final Answer Completion DetectionAUse when a task involves determining when an SDK session has completed a user-facing answer rather than intermediate tool activity to record the product, runtime and repository version, identity, trust boundary, data sensitivity, controlling artifact, owner, and approval scope before applying the method. Distinguish documented behavior from tested behavior, verify with a bounded reproducible case, preserve provenance, and report compatibility and uncertainty. Do not install, grant access, sen...toolsrustrailsVotes: 0GitHub stars: 2
Copilot Cli Hook Side Effect GovernanceAUse when a task involves reviewing automation hooks that run around Copilot CLI actions to record the product, runtime and repository version, identity, trust boundary, data sensitivity, controlling artifact, owner, and approval scope before applying the method. Distinguish documented behavior from tested behavior, verify with a bounded reproducible case, preserve provenance, and report compatibility and uncertainty. Do not install, grant access, send external writes, or change production pol...toolsrustgoVotes: 0GitHub stars: 2
Bot DevelopmentAUse when building or changing an automated bot that responds to messages, events, schedules, or platform APIs to define authorized actions, user consent, scopes, idempotency, rate limits, and failure behavior before connecting a live account. Trigger for chat, community, workflow, monitoring, or platform automation bots.toolsreactapiVotes: 0GitHub stars: 2
Agentic Loop DesignAUse when a request spans multiple actions or tool calls and needs a reliable completion condition to design a bounded agent loop connecting a clear objective to an artifact, feedback signal, allowed actions, durable progress, and stopping rules. Trigger for autonomous, recurring, or multi-turn work; use a simpler single pass for small, reversible tasks.toolsgoVotes: 0GitHub stars: 2
Pmtui SpawnAHand an independent part of your task to a child job — one headless agent run that does the task, reports a result and exits. Invoke when a separate agent could do that part without you, to dispatch it with one self-contained Message, poll its result, and act on the JSON.toolsgobashVotes: 0GitHub stars: 2
Agent Manager WorkerAThe persistent-heartbeat loop protocol for an agent-manager worker session. Invoke on every wake (the harness nudge says "continue per your agent-manager worker skill") to re-load the machine-report (WakeReport) schema and the operating rules — you reach the human only through the harness (your decision marker), you never decide when the project is done, and you report on EVERY wake with a fresh substantive status and next_step by overwriting your decision marker.toolsrustgoVotes: 0GitHub stars: 2
Agent Manager DeciderAThe reasoning protocol for an agent-manager supervisor consult — a decision typed policy found eligible for independent review, but has not approved. Decide whether and what the worker should do from the session goal, verify the claim with your read-only tools, and refuse whenever the goal or safety boundary does not clearly support the action.toolsrustgoVotes: 0GitHub stars: 2
Pmtui Ui TestingAHow to render and drive the pmtui dashboard under tmux to test the UI end-to-end — launch it headless on a scratch tmux server + scratch registry, navigate with send-keys, read the screen back with capture-pane, and exercise specific render states by seeding .project-state/ files. Invoke whenever you change anything the dashboard draws (src/bin/pmtui/, the row/preview/help/chrome render code, glyphs, keybar) and want to SEE the result instead of trusting a unit test, or when the user asks you...toolsrustgoVotes: 0GitHub stars: 2
Claude RepathAMigrate Claude Code state when a project folder is moved or renamed. Use this skill whenever the user mentions moving/renaming a project directory, reports that Claude Code has forgotten their project sessions/todos/memory after a folder change, finds broken ~/.claude/projects entries, asks "why did Claude forget my project" or "where did my history go", describes git worktrees with stale paths, or shows any symptom of lost session history / disappeared todos / missing context after folder re...toolspythongoVotes: 0GitHub stars: 4
Heinzel Macos CleanupBFind and remove leftovers of uninstalled apps on atoolspythongoVotes: 0GitHub stars: 96
ReleaseACut a Ponos release — pick the version, promote the changelog, tag it, and publish the GitHub release. Use when asked to release, ship, publish, cut a version, tag a version, bump the version, or prepare a changelog entry for a release.toolspythongoVotes: 0GitHub stars: 2
Eng ConsultAEngineering Consulting Panel — multi-voice review of code, plans, or architecture. Convenes a specialist roster, triages findings, returns a remediation plan.toolsrustgoVotes: 0GitHub stars: 3
reachowl-team-accountsASee which Facebook, Instagram, and WhatsApp accounts and browsers are connected to ReachOwl, pick the account to run campaigns from, and manage team members (invite, change role, remove). Use when the user asks which accounts they can use, why an account isn't available, or wants to add a teammate to ReachOwl.toolsapiVotes: 0
reachowl-queuesALook up and retry ReachOwl message queue items for a specific lead in a campaign. Use when a DM, friend request, or follow-up failed or is stuck, when the user asks whether a message was sent to someone, or wants to resend or reschedule a step.toolsVotes: 0
reachowl-profile-growthAGrow an Instagram profile with ReachOwl profile growth plans that auto-like and repost niche posts found by hashtag, keyword search, or Explore, on a daily limit and weekly schedule. Use when the user wants Instagram growth, engagement automation, or to create, pause, resume, or edit an Instagram growth plan.toolsgoreactVotes: 0
reachowl-contactsAList and inspect the leads (contacts) inside ReachOwl campaigns and add or delete CRM notes on them. Use when the user asks who is in a campaign, wants a lead's details or conversation notes, wants to log a note like 'replied yes', or needs to verify that an import worked.toolsapiVotes: 0
reachowl-comment-automationsAAuto-reply to Facebook and Instagram post comments with ReachOwl: when a comment contains a keyword (like 'price' or 'info'), post a public reply and send the commenter a private DM. Use for comment-to-DM funnels, lead magnets, giveaway posts, or to pause, edit, or check activity on a comment automation.toolsgoVotes: 0
reachowl-authAConnect and verify a ReachOwl account in the ReachOwl MCP server using the user's ReachOwl API token. Use when setting up ReachOwl, when any ReachOwl tool reports a missing or invalid token, when the user pastes a ReachOwl token, wants to log in with email and password, switch accounts, or log out.toolsgogitVotes: 0
reachowl-apiAMake a raw authenticated call to the ReachOwl V1 REST API through the ReachOwl MCP server when no named ReachOwl tool covers the request, such as a query parameter or field the named tools don't expose. Use only after checking the other ReachOwl skills.toolsexpressapiVotes: 0
cofounder-team-upgradeAUpgrade the cofounder-team skills to the latest version. Use when the user wants to update or upgrade their cofounder-team skills, get the newest versions of Jack, Maya, Priya, Dan, Sam, team, pitch-deck-coach, startup-application-coach, or humanizer, or asks to run /cofounder-team-upgrade. Triggers include 'upgrade cofounder-team', 'update my cofounder skills', 'pull latest cofounder-team', or running /cofounder-team-upgrade.toolsshellbashVotes: 0GitHub stars: 29
WizardAGenerate an interactive bash wizard walking a human through work only they can perform — credentials, secrets, one-off migrations. Use when provisioning infrastructure or third-party credentials, or when orc-env-provisioner escalates a human-only condition; never for steps the agent can do itself.toolsshellbashVotes: 0GitHub stars: 6
Verification Before CompletionAUse when about to claim work is complete, fixed, or passing, before committing or opening PRs — evidence before assertions. Doctrine skill (preloaded into orc-implementer), not a command — the bundled /verify covers interactive one-off behavioral checks.toolsrustgoVotes: 0GitHub stars: 6
Pr Size BudgetAUse when opening a PR or planning slices — defines the soft 300 LOC budget, the exclusion list, the shared gate prompt for ship/flow/stack-pr, and the override syntax.toolsgoshellVotes: 0GitHub stars: 6
Playwright QaADriver P of orc:browser-qa — browser QA with Playwright Test and the stock planner/generator/healer agents against a resolved QA target; yields a stitched step-tagged video, traces, and qa-manifest.json. Use when browser-qa selects the playwright driver; falls back to agent-browser.toolsgoshellVotes: 0GitHub stars: 6