verification

Plugin verification from marketplace pau-skills in paureis/pau-skills. Commands read from the repo's manifest — yesterday.

Skills (38)

• arena

  Build one task several ways at once: subagents with different briefs each make a version, a blind judge scores them, then pick a base and graft in the best parts of the rest. Use when the user says 'arena this', 'try it several ways and pick', 'design it twice', wants rival versions built and compar

• codebase-oracle

  Answer questions about the user's own project (how a part works, why it was built that way, whether something exists) from its code and history, not general knowledge, grading the evidence for each answer. Use when the user says 'answer these from the code', 'research this in the codebase', 'use the

• geppetto

  Designs, installs, audits and updates Claude Code subagents in .claude/agents, with a starter roster: eng-lead, builder, reviewer, writer, sweeper. Use when the user says 'create an agent', 'install the agent roster', 'audit my agents' or 'update my agents', or when agents overlap or carry leftover

• lean-subagent-dev

  Run the session as orchestrator: one subagent at a time builds each change from a written brief, you check its work and keep every push and merge, delegating to fit the usage budget. Use when the user says 'work lean', 'lean subagent', 'delegate this to a subagent', 'brief a subagent', wants subagen

• methodology

  Install a discipline for building software with AI agents in which no agent certifies its own work: contracts with verify clauses, a separate evaluator, two state files. Use when the user says 'adopt the methodology', 'set up a contract', or wants PROJECT.md and log.md, or when they ask how agents s

• overnight-run

  Leave an agent working alone on one task under a contract: trust checks first, a finish condition a script checks, an append-only log, reverts, an off switch and a morning audit by a fresh reviewer. Use when the user says 'run this overnight', 'keep going while I sleep', 'work on this while I am awa

• parallel-worktrees

  Get several independent tasks done at the same time: each goes to its own agent in its own git worktree and branch, with a check that no two edit the same files, then the branches merge back one at a time, testing after each. Use when the user says 'run these in parallel', 'split this across agents'

• dependency-security-audit

  Audit third-party dependencies and container images in any ecosystem for known vulnerabilities, fresh disclosures and supply-chain risk, and rank what to do. Use when the user says 'security audit', 'are we vulnerable', 'scan dependencies', 'CVE check', 'is this package safe', 'npm audit', 'pip audi

• dependency-upgrade

  Upgrade dependencies, runtimes and base images in any ecosystem, one major at a time or as a batch of minor and patch updates, each step tested and committed. Use when the user says 'upgrade dependencies', 'bump packages', 'upgrade to React 19', 'bump the Go version', 'update the base image', or whe

• rule-to-hook

  Turn one rule that keeps getting broken, not a new automation, into an enforced check (types, lint, CI, a test or a Claude Code hook) proven on a past mistake. Use when the user says 'enforce this rule', 'make this a hook', 'block this command the agent keeps running', 'stop the agent from doing X',

• safe-merge

  Merge a GitHub pull request without closing the pull requests stacked on it, by merging without branch deletion and deleting the head branch only if no open PR uses it as its base. Use whenever a PR is to be merged with gh, when the merge guard denied a gh pr merge, or when the user says 'merge PR',

• agent-friendly-audit

  Mine a project's history (reviews, fix commits, transcripts, instruction files) for mistakes agents repeat, and propose a check or design fix for each, backed by counts. Use when the user says 'make this repo agent friendly', 'find what agents keep breaking' or 'mine our history for agent mistakes',

• architect

  Sketch new code in the repository before building it: types, signatures and module boundaries in the project's language, two different shapes compared, then implement the deeper one test first. Use when the user says 'architect this', 'design this first', 'sketch the types', or before non-trivial co

• brainstorm

  Widen the options for an engineering or product decision in a codebase before settling it: options from named lenses, the agent recommends picks, code settles the facts, 3 or 4 rated. Use when the user says 'brainstorm this', 'give me alternatives' or 'widen this before we decide', or when a costly

• engineering-principles

  Named engineering principles to steer a design or code decision by: build the lever, model the domain, subtract before you add and more. Use when the user says 'apply build the lever' or 'which principle applies here', or before choosing a data shape, a new flag or type, or how to change many files.

• grill-me

  Interview the user one question at a time about a plan or design, with a recommended answer each and the code read for what it can answer, until every decision is settled. Use when the user says 'grill me', 'interrogate my plan', 'stress-test this design', 'poke holes in this', or wants to work thro

• improve-codebase-architecture

  Find deepening opportunities in existing code: refactors that turn shallow modules into deep ones, easier to test and to navigate, then talk the chosen one through. Use when the user says 'improve architecture', 'find refactoring opportunities', 'architecture review' or 'zoom out on the codebase', o

• sharpen

  Restate a task before doing it: the goal, the problem behind it, how an expert would approach it, what is missing, one outside check of the result; confirm in a few lines, then do the work. Use when the user says 'sharpen this', 'think before you start', 'make sure you understand the task', wants yo

• to-issues

  Break a plan, spec or PRD into independently grabbable issues, each a thin vertical slice through every layer, ordered by dependency and filed after approval. Use when the user says 'break this into tasks', 'create issues', 'create tickets', 'make tasks from this', 'split this into slices' or 'break

• to-prd

  Write a PRD (product requirements document) from what the conversation and the code already settled, without interviewing, and save it as a markdown file. Use when the user says 'write a PRD', 'make this into a PRD', or wants an agreed feature, plan or finished grilling session written up as a spec

• handoff

  Write a handoff so a fresh session or another tool can resume with no prior context, as HANDOFF.md or a block to copy. Use when the user says 'handoff', 'hand off', 'pause here' or 'wrap up for now', or when work will resume later or the session is getting long. Pauses work in progress; does not com

• instructions-audit

  Audit existing agent instruction files (CLAUDE.md, AGENTS.md, rules) for token cost, duplicates, contradictions, stale paths, vague rules and secrets, and propose edits. Use when the user says 'audit my CLAUDE.md' or 'clean up CLAUDE.md', or when CLAUDE.md is too long, the agent keeps ignoring its r

• project-setup

  Set up Claude Code for a project through a short interview for newcomers, in their language, then write CLAUDE.md on a yes and start a first task. Use when the user says 'set up Claude Code for this project', 'help me get started with Claude Code' or 'onboard me', or when the user is new to Claude C

• retrospective

  Review this session and turn its corrections and redone work into short proposed edits or checks for CLAUDE.md, memory or a skill, checked against what is already written. Use when the user says 'retrospective', 'what did we learn', 'how should skills be updated', asks what to do differently next ti

• session-close

  End a working session in a fixed order: clean tree, short retrospective, roadmap and log, handoff file, commit and push. Use when the user says 'close the session', 'end the session' or 'wrap up this session', or when work stops for the day, the context is nearly full, or a roadmap item is done and

• which-skill

  Recommend which pau-skills skill, or sequence of skills, fits the user's situation, what the installed hooks already do, and which skills the user actually starts. Use when the user asks 'which skill should I use', 'what can these skills do', 'where do I start', 'is there a skill for this', 'help me

• bug-fix

  Debug and fix a bug at its root cause from runtime evidence: reproduce it twice, test one hypothesis at a time, commit a failing test first. Use when something is broken and the cause is unclear, a first fix did not hold or the bug came back, or for 'find the root cause', 'why does X return Y'. An o

• ci-cost-and-cadence

  Decide how much CI to run and when, weighing minutes, money and waiting time against what each check has caught, then build the choice with safeguards. Use when the user says 'run tests once per PR', 'move tests to a nightly run', or when CI minutes or the bill run out, PR feedback is slow, someone

• evaluator

  Check a finished change against its contract, issue or acceptance criteria: a fresh agent that has not seen the build attacks it, grading each assertion PASS, FAIL or UNVERIFIED with reproducible evidence. Use before calling an increment done, or when the user says 'evaluate this', 'run the evaluato

• flaky-test-hunt

  Find why a test or CI job fails only sometimes instead of retrying it: measure the failure rate, find and fix the cause, prove the fix with enough clean runs. Any language. Use when the user says 'this test is flaky', 'fails intermittently', 'passes locally but fails in CI', 'can we just retry it',

• hillclimb

  Improve one number of the project (run time, bundle size, memory, a score) in measured steps: one change per attempt, kept only when it beats the noise, each kept win committed. Use when the user says 'make this faster until', 'get the bundle under 200 KB', 'push the score up', 'optimize this metric

• launch-readiness

  Give an honest go or no-go for the stage an app is entering (beta, paying customers, enterprise, regulated data), backed by evidence and probes, in a READINESS.md ledger that rechecks what changed. Use when the user says 'is this ready for paying customers', 'check launch readiness', 'can we launch'

• migration-review

  Review a schema or data migration (any tool or raw SQL) before it runs: locks, downtime, deploy compatibility, data loss, reversibility, run time; gives a verdict and safe rewrites. Use for any change under a migrations folder, or 'review this migration', 'is this migration safe', 'will this lock th

• mutation-test

  Prove a test or guard can fail: introduce the exact defect it claims to catch, run it, then restore the file. Use after writing a test for a bug or a security rule, or when the user says 'mutation test', 'can this test fail', 'is this test any good', 'could this test catch a bug', 'prove the test ca

• review-triage

  Decide which review findings are worth fixing, fix those with a test each, and end the review and fix loop after one round. Use when the user pastes review findings or says 'triage these review comments', 'is this worth fixing', 'the review keeps finding things', 'what blocks the merge'. Sorts a rev

• skill-eval

  Check blind whether a change to a skill or prompt changes what an agent does: headless runs per variant, scored by a reviewer who does not know which is which; or test which skill starts for which request. Use when the user says 'A/B this prompt', asks whether a change to a skill or prompt helped, o

• tdd

  Test-driven development: build or fix code one behaviour at a time in a red-green-refactor loop. Use when the user says 'test first', 'TDD', 'test-driven', 'red green refactor', 'write tests', 'build this with tests', 'add test coverage', 'make sure this is tested'. Writes and runs the tests, not a

• verify-app

  Write or maintain the project's own verification skill and feature map, so any agent can launch the real app, drive it like a user and keep evidence. Use when the user says 'make a verification skill', 'maintain the verify skill', 'match the skeleton', or asks how an agent can test the app, or when

Similar plugins