mattpocock-skills
claudeMatt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Sift — Manage the Connect roadmap and backlogs Built on the canopy agent operating model: a turn-driven persona with reads-free / writes-gated guarding and human-approved outbound actions.
Not in a marketplace yet, so it loads for one session with --plugin-dir.
Plugin sift in dimagi-internal/sift. Commands read from the repo's manifest — 5 days ago.
agent-turn-reviewSift's pre-send review — invokes the fleet-wide `canopy:agent-turn-review` discipline, then adds Sift's specifics. Run before EVERY outbound reply / deliverable / PR.
answer-callerAnswer a caller (not Sift's owner or an admin) in a canopy-confined session.
capacity-refreshRefresh Sift's picture of Connect team capacity, throughput and velocity from the Connect Tech Sprint Performance dashboard, and rewrite context/capacity-snapshot.md. Use when the snapshot is over 14 days old, when a new sprint has started since its date, before a capacity/prioritization answer, or
clean-the-backlogClean the backlog — Sift's weekly sweep of the Connect delivery backlog (CCCT): find stale, unowned, duplicate, under-specified or mis-classified (component / team / epic / roadmap link) tickets and CCCT work on parked roadmap items, turn each finding into a concrete proposed change, get the batch a
grooming-prepPrepare Connect's Thursday grooming session for ONE sprint at a time, focused on ticket quality. Starts by listing the upcoming sprints across the whole CCCT project and asking which one to prepare. Then checks every planned ticket in that sprint for a clear description, acceptance criteria (or the
interrupt-summaryInterrupt Summary — Sift's monthly read-out of what lands on the CI (Connect Interrupt) board: tickets per month by type (task / bug / story), what each category of request is, and the product change that would take it off the board. Rebuilds the shared "Connect Interrupt Summary" artifact page for
manager-syncSift's periodic advisor sync — reviews everything done since the last sync, HARSHLY self-grades the work AND the reusable skills built, and puts a short list of already-decided next steps to the advisor to confirm or redirect. The canonical procedure is fleet-wide and lives in the installed canopy p
product-principlesRead, apply and maintain the Connect Product Principles — the Confluence page of product principles and decisions that have been made (payments, visits, FLW communication, support, language). Use it to check roadmap items and tickets against settled decisions, and to draft and apply updates to the p
review-interruptReview Interrupt — Sift's weekly pass over the CI (Connect Interrupt) board, where ad hoc support issues are reported. Finds interrupts that nobody owns, that stopped moving, that have been open too long or that don't belong on the board, decides per ticket whether to nudge, flag or re-home it, and
review-stale-ticketsReview Stale Tickets — sweep ONE Connect Jira project for tickets that have stopped moving (CCCT = implementation/delivery, CCC = roadmap), decide per ticket whether to close, nudge or leave it, and get the batch approved before bin/sift-jira applies it. Use when asked to "review stale tickets", "cl
roadmap-intakeSift's intake triage for NEW items in the CCC roadmap backlog — check every idea created since the last run for the intake basics (Prioritization, Goal, a good description, 1–2 Product Areas, and Product Pod once prioritized), propose the missing values and a drafted problem statement, get the batch
setupSet up this machine to run Sift: bind this install to ONE agent instance on canopy-web and fill ~/.sift/.env from that instance's Canopy vault. Use on first run, a new machine, or a missing/stale .env.
shippingHow Sift ships code — branch, PR, wait, merge, verify it landed — and specifically how to WAIT for a PR without burning the turn on a hand-rolled sleep loop. The canonical procedure is fleet-wide and lives in the installed canopy plugin (agent-core/shipping.md); this stub binds it to Sift. Use whene
sprint-reportWrite the summary report for a Connect sprint that has just closed in Jira: what was completed, what carried over, the work grouped by the 2027 goals it serves, which epics finished, and the story / task / bug split. One pod per report, kept short. Use when a sprint closes (on Sift's schedule it che
task-trackerSift's project/task state — one board task per iterative thread/project, backed by canopy-web (kanban at /agents/sift). The canonical procedure is fleet-wide and lives in the installed canopy plugin (agent-core/task-tracker.md); this stub binds it to Sift. Use when taking on multi-turn work, when a
turnSift's full turn-of-work orchestrator. Use when a human says "do a turn", "check your inbox", or otherwise triggers Sift to work. The canonical procedure is fleet-wide and lives in the installed canopy plugin (agent-core/turn.md); this stub binds it to Sift's identity. THE entry point for a turn.
Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).
39 skills
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.
1 MCP server
Automates browser interactions for web testing, form filling, screenshots, and data extraction
12 skills
Orchestrate multi-agent teams for parallel code review, hypothesis-driven debugging, and coordinated feature development using Claude Code's Agent Teams
6 skills · 7 commands · 4 agents
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.