Plan and safely run one or more canonical issue, existing PR, or durably overridden ad-hoc work lanes with coordinated subagents, validation, review, and merge-readiness. Unbound direct prompts route through planning/reconciliation before implementation…
shakacode/agent-workflows
- Latest recorded source activity
- SkillsMP catalog refreshed
- skills collected
- 36
- GitHub stars
- 8
- GitHub forks
- 1
Install with an AI assistant
Copy this prompt into the AI assistant you're using.
Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/shakacode/agent-workflows.Loading the repository README…
Skills in this repository
Showing 36 of 36 collected skills.
Use when auditing merged PRs after concurrent agent work, before a release candidate, after a suspected bad merge, or when checking for missed reviews, missing changelog entries, cross-PR interactions, or release risk.
Use when a test fails intermittently in CI — classify the intermittency, work from the real CI error, fix the root cause systemically, and verify with a green CI build rather than local runs.
Use when local validation is green but hosted CI is red, a CI-only failure needs reproduction, or runner/toolchain parity is suspected.
Fetch GitHub PR review comments, triage them into must-fix/discuss/optional/skipped, and guide fixing or replying to selected feedback. Use when addressing PR review comments or review threads.
Use when verifying changed behavior in a real running app or service with recorded HTTP, browser, or CLI evidence, including acceptance criteria and cheap unhappy paths.
Use when preparing a ready prompt for Claude, Codex, or another agent to perform a review-only GitHub issue triage or open-issue audit, especially broad scans such as all current issues, unlabeled issues, performance-regression clusters, close candidates, or…
Manually verify that a bug-fix PR actually works by reproducing the failure before the fix and confirming it is gone after, with captured evidence, then posting findings to the PR (and optionally the linked issue). Use when asked to manually verify a PR/fix,…
Use when choosing GitHub issues or PRs for a PR batch, recommending and grouping worker lanes by model/reasoning-effort assignment, preparing a subagent batch plan, or producing a ready goal prompt that invokes pr-batch.
Explain a pull request as a conceptual walkthrough, either interactively one change at a time or as one complete COMMENT-only GitHub review when explicitly requested. Use when a user asks to understand, explain, present, tour, or walk through a PR or diff;…
Generate a whole-surface issue/PR inventory, dependency graph, and capacity-aware pr-batch split from live GitHub plus coordination-backend state. Use when a coordinator needs current issue/PR inventory and batch planning.
Use when asked where a dispatched PR batch stands - mid-flight or after dispatch - to report per-lane coordination and live GitHub state in the canonical readiness vocabulary. Read-only; use close-batch for completed-batch closeout and pr-monitoring for a…
Use when a PR needs skeptical pre-merge or post-merge risk review, especially after concurrent agent work, before merge readiness, before a release candidate, or when Codex or Claude should red-team correctness, security, compatibility, changelog, validation,…
Run a structured second-model code review as a closeout gate on a local, branch, or commit diff, then verify every finding against the real code and loop until clean. Use before commit/push/ship in this repo.
Close an active agent task with live-state verification, durable outcome capture, explicit follow-up ownership, and an archive-readiness verdict. Use when the user says "close session", asks "anything else pending?", "any decisions needed?", or "should we…
Resume an in-progress task with a structured checkpoint instead of a bare "continue", re-establishing what is done, what is next, and how done is verified. Use when resuming work after an interruption, handoff, or a vague "keep going".
Print restart-safe copy-paste prompts for pausing an agent thread before restarting Codex, Claude, or another runner and for resuming in the same thread or a new chat from the handoff. Use when the user asks for a pause prompt, restart prompt,…
Use when reviewing an implementation plan before coding or launching workers to check approach, scope, and repo-convention fit.
Use when monitoring an opened pull request through current-head checks, comments, conflicts, merge-readiness, and final handoff.
Prepare or recover Codex tasks and subagents around an app restart, crash, or account change. Use when preparing for a restart or recovering from checkpoints, logs, and live state when no handoff exists.
Optional meta-maintenance skill for recording sanitized observations that may improve shared skills or durable workflow lessons. Use only when the user explicitly asks for task observation, skill improvement capture, or observation review; do not activate it…
Use before fixing, batching, or assigning GitHub issues or proposed fixes when the value is uncertain, the report came from AI/code analysis, the fix is complex, or the user asks whether an issue is worth doing. Produces an evidence-backed recommendation: fix…
Recover and close stale or unfinished PR-batch tasks by resolving lifecycle ownership, continuing canonical closeout, walking the maintainer through current PRs or blockers, and archiving only after the archive gate passes. Use when a batch task has lingered,…
Analyze current branch changes with the repo CI detector and run user-selected local CI jobs. Use when the user asks to run, reproduce, or choose local CI checks.
Run a local verification loop for the current branch before creating or updating a PR, selecting checks from the repo's binstub contract and changed files. Use when asked to verify, test, or prepare PR changes.
Audit multiple Codex tasks and visible ChatGPT chats for archive and follow-up actions, or recover interrupted Codex tasks when explicitly requested. Use for machine-wide cleanup and recovery after quota exhaustion or app restarts; use close-session or…
Analyze merged PRs and update the changelog, optionally stamping release, rc, beta, or explicit version headers. Use before releases or when changelog entries are missing.
Drive a portable red-green-refactor workflow for features, bug fixes, and changes to existing behavior. Use when implementing behavior with test-driven development, reproducing a bug as a failing test before fixing it, changing the semantics of existing code,…
Safely assess an outside-contributor fork pull request through metadata and diff evidence, then report a maintainer decision without executing untrusted content. Use when screening an outside-contributor fork pull request.
Use when a diff is correct but may still be making the codebase worse — file growth, scattered conditionals, thin abstractions, layer violations, or feature-flag branching debt.
Use when an implementation request is vague, high-ambiguity, or needs requirements, design, and executable tasks before plan-pr-batch or pr-batch.
Audit GitHub Actions workflows and composite actions for expression injection, broad reusable-workflow secret inheritance, mutable or undocumented external references, and actions outside a closed trusted allowlist. Use when assessing GitHub Actions trust…
Use when explicitly asked to run destructive QA stress testing against repo-owned demo or target apps, especially for leakage, memory growth, performance degradation, hostile inputs, fault injection, pentesting, and gated finding reports.
Report tight progress on the current work - done, in progress, blocked, and next - without starting new work. Use when asked for a status update or "where are we / anything needed from me".
Use when verifying performance-sensitive changes with baseline-vs-patched benchmark evidence, repeated runs, noise-aware verdicts, and repo-seam benchmark commands.
Use when reviewing data models, signatures, domain types, parsing, casts, or state machines so invalid states stay hard to represent.