sdlc-skills

Full SDLC toolkit — all agents and skills in one install.

31 skills · hooks

Chrome DevTools Protocol browser automation — inspect, click, screenshot, verify UI.

1 skill

End-to-end bugfix workflow: reproduce → root cause → fix → regression test.

1 skill

Record a problem → outcome → lesson when a hypothesis closes or an experiment concludes.

1 skill

Structured code review for correctness, security, performance, and maintainability.

1 skill

Five-step task handoff protocol: commit → push → PR → comment → notify.

1 skill

Checkpointed research workflow — trend research, topic analysis, fact-checking.

1 skill

Draft, stress-test, and ratify measurable outcome anchors (dated baseline + target + timeframe).

1 skill

Author and maintain canonical persona cards (surface, goals, pains, JTBD, evidence) referenced by stable slug.

1 skill

Read-only discovery pipeline dashboard — gate state, what's blocked and on whom, the exact next action per item.

1 skill

Per-project/session/role/day and per-sub-agent AI-agent cost & efficiency, metered by ccusage — separates orchestrator overhead from dispatched sub-agents (model-aware); snapshots + before/after diffs.

1 skill

Cross-channel context gathering — local KB, email, Teams, web — before responding.

1 skill

Git operations: branching, commits, PRs, rebasing, cherry-picks, releases.

1 skill

Socratic one-question-at-a-time stress-test of a decision or hypothesis, capturing outcomes inline (DEC records, edits).

1 skill

Feature implementation workflow: plan review, tests, implementation, verification, delivery.

1 skill

Operational health, alert investigation and monitoring handover.

1 skill

Front door for raw asks — verdicts each Act Now / Plan Next / Collect More Signal / Decline-or-Defer and mints in-scope items as Problems.

1 skill

Create and manage issues in GitHub, Linear, or GitLab.

1 skill

Convergence pass — classify each journey COVERED / GAP / OUT-OF-SCOPE and author missing Problem + Hypothesis stubs.

1 skill

Per-role persistent memory — facts, preferences, decisions, and daily log as plain markdown.

1 skill

Microsoft 365 integration — email, Teams, calendar, SharePoint via Graph API.

1 skill

Manual mobile-app testing via Appium / device-farm sessions — native local plus Mobitru cloud real devices.

1 skill

Headless Obsidian vault operations — create, search, and file notes without the app.

1 skill

Maintain the opportunity–solution tree over existing artifacts — the disguise test + Olsen scoring, regenerates the outcome-tree board.

1 skill

Structured feature planning: requirements, investigation, feasibility, implementation plan.

1 skill

UI/E2E test automation with Playwright MCP.

1 skill

Evidence-based pre-production readiness and release gates.

1 skill

Rank incubating/promotable hypotheses (RICE default; WSJF/ICE options); confidence derived from each bet's evidence band.

1 skill

Staged rollout, promotion and interrupted-run reconciliation.

1 skill

Release plans, artifact identity, authority and chat-approved exceptions.

1 skill

Compatible rollback, roll-forward and verified recovery.

1 skill

Append-only residual-risk register — records, accepts, expires, closes, supersedes and tickets risks via register.mjs.

1 skill

Evidence-gated secure code review with re-checkable file:line citations — findings, second opinions, and fix verification via cite.mjs and verify.mjs.

1 skill

Orchestrates a security engagement end to end for the security-lead — init, review, register, model, cases, hand-off, report, fix, acceptances, sign-off.

1 skill

Turns a threat model's candidate cases into passive security test cases in manual-qa format, admits them by lint, and hands off a verified suite via cases.mjs.

1 skill

Generate AGENTS.md, .agents/ project docs, and per-role memory briefings after codebase exploration.

1 skill

Distill past Claude Code sessions (main + sub-agent) into evidenced, ack-gated memory and workflow improvements (used by scout).

1 skill

Prepare a value-of-information-ordered interview guide, or synthesize raw notes into evidence and propagate answers into hypotheses.

1 skill

End-to-end test automation workflow: explore → specify → implement → review. Pluggable TMS adapters (Zephyr Scale, TestRail, Xray, Azure Test Plans, markdown).

1 skill

Execute a TMS test case end-to-end, capture stable selectors, flag defects, and emit an Automation-Friendly Spec (AFS). Used by qa-engineer. Does not write automation code.

1 skill

Code-derived data-flow diagram and STRIDE threat model with a citation per element and mitigations recorded as claims a fresh reviewer can confirm or refute.

1 skill

Goal-backward verification — checks whether the desired outcome was actually achieved.

1 skill

Bootstrap, configure, and author tests for the Vividus BDD framework — JBehave-based .story files with 47+ plugins (web, REST, mobile, DB, messaging, AWS/Azure, visual, accessibility), BOM-pinned versions, suite/profile/environment configuration.

1 skill

Read .xlsx/.xls spreadsheets (test cases, checklists, requirement matrices) into Markdown tables for agent ingestion.

1 skill

Android/Kotlin developer agent — greenfield Jetpack Compose, Room/DataStore, Gradle/AGP with TDD and no-emulator discipline.

Manual-QA onboarding agent — profiles a web app and writes the shared app_profile.md that every manual-qa agent reads.

Business analyst agent — turns vague ideas into user stories with acceptance criteria.

Discovery research agent (Sam) — stakeholder interviews, market/desk research, and adversarial verification of hypotheses; dispatched by product-owner.

iOS/Swift developer agent — SwiftUI, SwiftData, Swift Testing with TDD and no-simulator discipline.

JavaScript/TypeScript developer agent — React, Next.js, Node services with TDD.

Personal assistant agent — email, calendar, Teams, notes, and daily briefings.

Product Owner discovery-lead agent (Priya) — runs intake triage, persona/outcome framing, opportunity-tree mapping, and prioritization; guards the promotion gate.

Project manager agent — routes tasks to developers, coordinates team progress, owns the merge gate.

Python developer agent — implements Django, FastAPI, Flask services with TDD.

QA engineer agent — verifies features, reproduces bugs, writes E2E tests via Playwright.

Dora — release readiness, staged rollout, recovery and infrastructure health.

Codebase scout agent — maps unfamiliar repos, generates CLAUDE.md and project configuration.

Security lead agent (Rasmus) — orchestrates a threat-led security engagement: dispatches the threat-modeler and security-reviewer, checks their citations, admits passive test cases, writes and signs off the assessment, verifies fixes, and proposes acceptances. The only human-facing security role.

Security reviewer agent (Vera) — evidence-gated secure code review: review, vulnerability-review, mitigation-review, and fix-review, each a fresh dispatch with re-checkable citations.

Tech lead agent — decomposes stories into dependency-ordered tasks, owns architecture decisions and code review.

Manual-QA authoring agent — turns rough ideas, bug reports, or stories into formatted TC-NNN cases.

Test automation engineer agent — turns automation-ready specs into green tests inside the project's existing framework (Playwright, Cypress, pytest, JUnit, NUnit, WDIO).

Manual-QA reporting agent — turns runner results into a Markdown run report.

Manual-QA run orchestrator — assembles the suite, dispatches a runner per case, triggers the report.

Manual-QA execution agent — runs one case against the running app via Playwright MCP and returns a structured JSON result.

Manual-QA sizing agent — rates case size/complexity (S/M/L) before authoring and flags Large ones to split.

Threat modeler agent (Ilse) — derives a code-based data-flow diagram and STRIDE threats with a citation per element, records mitigations as claims, and assigns one disposition per threat.