pell

Plugin pell from marketplace pell-skills in FroncekB/pell_skills. Commands read from the repo's manifest — 1 month ago.

Skills (8)

• autonomous-build-commit-essays

  Use when writing commit messages during greenfield product-building sessions, multi-commit feature series, or any session that will produce more than 5 commits. Also use when the commit history is itself serving as the product's design narrative (e.g., an investor will read `git log`, or the session

• autonomous-build-jealousy-ranking

  Use when ranking a feature backlog under a "next level / take it further / mad jelly" mandate, when writing a product roadmap for someone who wants differentiation (investors, launch partners, review threads), or when asked to file GitHub issues for a list of candidate features.

• autonomous-build-purpose-layers

  Use when building or extending a product in an autonomous multi-hour session with an ambiguous "go big / next level / use the time well" mandate, and you could build either shallow-horizontal (more features of the same kind) or deep-vertical (each commit deepens the product's reach into a new semant

• autonomous-build-session-pacing

  Use when the user grants an autonomous multi-hour budget ("4 hours", "keep going until I stop you", "you have 2.6h left"), or when the user asks you to check in on the clock, or when working quiet stretches with no per-change user feedback.

• autonomous-build

  Use ONLY when the user explicitly names this skill — e.g., "use autonomous-build", "kick off with autonomous-build", or an equivalent explicit request. Do NOT auto-fire on generic "build me X" / "new project" / "take it further" phrasings; those go through the normal flow.

• coordinate-agents

  Use when a task is large enough to split across multiple subagents that you coordinate as the lead, rather than doing it all yourself in one context.

• fan-and-critic

  Use when working a long autonomous build or design session with little per-change user feedback, and you want grounded outside eyes on each change without the user having to review everything themselves.

• frontend-router

  Use at the START of any user-facing UI work in a Pell repo — building React components, pages, layouts, design systems, styling, Tailwind/CSS, marketing pages, dashboards, forms, or any task whose deliverable is something a human will look at in a browser.

Slash commands (22)

• /pell:address-review

  Pull the inline + general comments off one of your Bitbucket PRs and triage each — apply mechanical fixes to your working tree or reply on the thread. Never commits, never pushes, never resolves threads.

• /pell:correctness-review

  Review a code change for correctness only (logic errors, broken invariants, missing error handling). Works on a Bitbucket PR or local changes. Returns a markdown report with all findings.

• /pell:finish-work

  Close out a branch by opening a Bitbucket PR and (only on explicit consent) transitioning the linked Jira ticket and adding a PR-link comment. Read-only against Jira by default; PR creation always confirmed.

• /pell:four-pass-review

  Four-pass review of a Bitbucket PR — dispatches correctness, quality, security, and test-coverage reviewers in parallel with linked Jira context; add "skip tests" to drop the test-coverage pass. Offers a run-marker comment on the PR (default yes), then optionally posts findings as inline PR comments

• /pell:from-ticket

  Compose ticket-to-implementation in one command. Fetches a Jira ticket, runs the /pell:scope readiness check, dispatches /pell:start-work to create a branch, then hands off to superpowers:brainstorming → writing-plans for the design and plan. When superpowers isn't installed, runs a lightweight inli

• /pell:groom

  Walk a set of Jira tickets (an epic's open children, a sprint, a JQL query or key list, or the backlog), map the code they touch, and review each as a technical architect and business analyst — requirement gaps, code conflicts, blast radius, cross-ticket collisions, better approaches, and missed bus

• /pell:local-review

  Four-pass review of local uncommitted changes — dispatches correctness, quality, security, and test-coverage reviewers that respect CLAUDE.md and surrounding code conventions; add "skip tests" to drop the test-coverage pass. Optionally applies suggested fixes.

• /pell:map-repo

  Map a repo's Jira, Confluence, Drive, and Bitbucket coordinates into a committed docs/pell/context.md so every session stops re-deriving them, and optionally build the project SOW at setup time. Read-only against every remote system; all writes are local and gated.

• /pell:my-tickets

  List Jira tickets assigned to you that aren't done yet. Optional freeform filters (project key, status). After rendering, offers to chain into /pell:start-work for any ticket you pick.

• /pell:precheck

  Check whether a ticket is already filed, in progress, or shipped — searches similar Jira tickets, repo implementation, in-flight PRs/branches, and recently-merged commits, then renders a verdict. Read-only by default; offers a gated duplicate-link/comment only when run against an existing ticket key

• /pell:quality-review

  Review a code change for code quality only (readability, naming, duplication, convention adherence). Works on a Bitbucket PR or local changes. Returns a markdown report with all findings including nits.

• /pell:related

  Show everything connected to a Jira ticket — linked issues (blocks, relates to, duplicates), parent/subtasks, external links (PRs, docs), and Bitbucket PRs referencing the key. Auto-detects the key from the current branch when run with no arguments. Read-only.

• /pell:repo-review

  Whole-repo code-quality audit — walks the codebase, dispatches repo-quality-reviewer agents in parallel, aggregates findings. Looks for duplicated logic, dead code, convention drift, oversized files, tight coupling, ignored warnings. Read-only.

• /pell:repo-security-review

  Whole-repo security audit — walks the codebase, dispatches repo-security-reviewer agents in parallel, aggregates findings. Two passes per file: regex scan for sensitive data (SSNs, credit cards, API keys, JWTs, private keys), then code-level vulnerability review (XSS, SQLi, path traversal, crypto mi

• /pell:review-queue

  List open Bitbucket PRs where you're a requested reviewer (one repo, several, or the whole workspace), then chain into a review on the one you pick. Read-only.

• /pell:scope

  Place a Jira ticket in the whole project — offers to build and cache a synthesized Statement of Work from the project's epics and linked Confluence or Drive scope docs (checking for an existing SOW first), then runs a Definition-of-Ready check on the ticket (description, acceptance criteria, epic fi

• /pell:security-review

  Review a code change for security issues only (injection, authn/authz, secrets, OWASP top-10). Works on a Bitbucket PR or local changes. Returns a markdown report with all findings including low-severity observations.

• /pell:start-work

  Fetch a Jira ticket, create a properly-named local branch, and (only on explicit consent) assign or transition the ticket. Read-only against Jira by default — side-effects are opt-in per action or pre-authorized inline ("assign to me", "move it to in-progress").

• /pell:test-review

  Review a code change for test adequacy only (untested behavior, tests that can't fail, missing edge/error coverage, flaky patterns). Works on a Bitbucket PR or local changes. Returns a markdown report with all findings.

• /pell:three-pass-review

  Deprecated alias for /pell:four-pass-review — the PR composite was renamed when the test-coverage pass became default-on. Prints a one-line notice, then forwards all arguments to /pell:four-pass-review unchanged. Will be removed in a future release.

• /pell:triage

  List unassigned Jira tickets in a project so you can sort through the queue. Groups by priority, then per ticket prompts whether to claim it (assign to you), start work on it, or skip. Read-only unless you say yes per ticket.

• /pell:wrap-up

  Close out a branch in one command — runs /pell:local-review on the working tree, offers to commit any fixes, then dispatches /pell:finish-work to push, open the PR, and transition Jira. Read-only against Jira by default; all side-effects gated.

Subagents (15)

• conductor-correctness-reviewer

  Use to adversarially review a code or doc change for correctness before merge — one of a dual-model pair.

• conductor-design-steward

  Use to review a UI surface for visual and verbal cohesion with the rest of the product — tokens, primitives, copy voice, and fidelity — not correctness.

• conductor-implementer

  Use to build and wire a single delegated unit of work end-to-end in an isolated worktree — implementation, not review.

• conductor-integration-gap-auditor

  Use to check whether a built change is actually connected — wired to real callers, endpoints, and producers — rather than merely correct or compiling; one of a dual-auditor pair.

• conductor-scout

  Use to investigate a question about the system — a drift hunt, reinvention audit, root-cause analysis, or area map — with grounded evidence before a build decision depends on the answer.

• correctness-reviewer

  Reviews a code change for correctness — logic errors, off-by-one bugs, broken invariants, missing error handling at real boundaries, race conditions, regressions, and mismatches with stated intent (Jira ticket or CLAUDE.md). Returns ALL findings including nits. Use as part of /pell:correctness-revie

• quality-reviewer

  Reviews a code change for code quality — readability, naming, duplication, premature abstraction, dead code, and especially adherence to repo conventions (CLAUDE.md, linter configs, surrounding patterns). Returns ALL findings including nits. Use as part of /pell:quality-review, /pell:four-pass-revie

• repo-mapper

  Walks a checkout and its linked Jira, Confluence, Drive, and Bitbucket surfaces to produce the repo's coordinates — project keys, exact transition names, space and page ids, folder ids, repo slug. Read-only; writes no files and asks no questions. Dispatched by /pell:map-repo.

• repo-quality-reviewer

  Reviews a repository (not a diff) for code quality, reusability opportunities, dead code, convention drift, and files that have grown too big. Operates on a focused list of files supplied by the dispatching command. Returns ALL findings including nits. Use as part of /pell:repo-review.

• repo-security-reviewer

  Reviews a repository (not a diff) for sensitive data (SSNs, credit cards, API keys, JWTs, private keys, driver's license patterns) and code-level vulnerabilities (XSS, SQLi, path traversal, hardcoded credentials, crypto misuse, PII logging). Operates on a focused list of files supplied by the dispat

• security-reviewer

  Reviews a code change for security vulnerabilities — injection, authn/authz gaps, secret leakage, unsafe deserialization, missing input validation at trust boundaries, OWASP top-10. Returns ALL findings including low-severity nits. Use as part of /pell:security-review, /pell:four-pass-review, or /pe

• sow-builder

  Walks every issue in a Jira project plus any linked Confluence or Google Drive scope documents and synthesizes a per-epic Statement of Work (scope statements, deliverables, dependencies, gaps) as a markdown document. Dispatched by /pell:scope when the SOW cache is missing or stale and by /pell:map-r

• test-reviewer

  Reviews a code change for test adequacy — whether new or changed behavior is tested, whether the tests actually assert that behavior (not tautological or mock-only), coverage of edge and error paths, and flaky-test patterns. Returns ALL findings including nits. Use as part of /pell:test-review, /pel

• ticket-code-auditor

  Audits one Jira ticket against the code it would touch, as a technical architect and a business analyst — requirement gaps, code conflicts, blast radius, collisions with neighboring tickets, better approaches that extend existing code, and missed business requirements — grounding every finding in fi

• ticket-code-mapper

  Maps the code a cluster of Jira tickets would touch — entry points, files, state, rules, consumers, and data per flow — into one shared area map. Read-only; judges nothing. Dispatched by /pell:groom.

Similar plugins