preetamnath/agent-skills

1

durable-docs-update After coding work, sweep comments in changed files and sync only related durable guidance through WORTH → PLACE → SHA…

preetamnath/agent-skills

39

2

execute-plan Implement a feature by executing its wave-grouped plan.md. TRIGGER when: user says 'run the plan', 'execute the plan'…

preetamnath/agent-skills

29

3

product-interview Define WHAT to build — product scope and UX — before technical design. TRIGGER when: user asks for a product intervie…

preetamnath/agent-skills

24

4

tech-design Turn a locked product/UX spec into technical design — the HOW: architecture, data shapes, signatures, file layout. TR…

preetamnath/agent-skills

23

5

write-plan Sequence a locked spec into dependency-ordered, wave-grouped tasks. TRIGGER when: user says 'build the plan' or 'sequ…

preetamnath/agent-skills

22

6

interview-me Socratically interview the user to move an open question — a decision, strategy, trade-off, or research direction — f…

preetamnath/agent-skills

20

7

refine-file Audit one current-guidance instruction file for worth, placement, and clarity.

preetamnath/agent-skills

20

8

execute-chat Execute work already agreed in the current chat without a spec or plan.md. TRIGGER when: user says 'ready to execute'…

preetamnath/agent-skills

19

9

shopify-dev-mcp Routes Shopify Dev MCP lookups for surfaces the bundled Shopify skills don't cover. TRIGGER when: user needs docs or …

preetamnath/agent-skills

18

10

grill-me Stress-test a plan, design, or decision — challenge assumptions, expose gaps, force specificity. TRIGGER when: user s…

preetamnath/agent-skills

17

11

tighten-file File-level tightening pass on an instruction file (CLAUDE.md, skill, agent prompt, style guide) using `tighten-instru…

preetamnath/agent-skills

17

12

two-pass-review Two-pass code review — a review pass hardened by an adversarial verification pass that suppresses false positives. TR…

preetamnath/agent-skills

17

13

fix-verify-loop Resolve confirmed P0/P1 findings in at most two fix-and-verify attempts, then escalate. TRIGGER when: review findings…

preetamnath/agent-skills

15

14

plan-runner Executes wave-grouped markdown plans via parallel subagents. Orchestrates implementation, per-wave review, fix cycles…

preetamnath/agent-skills

15

15

sentry-analysis Diagnose the root cause of a Sentry issue from its error logs, breadcrumbs, and codebase context.

preetamnath/agent-skills

15

16

explain-simply Re-explain your last response in the simplest language. TRIGGER when: user says 'explain simply', 'explain in simpler…

preetamnath/agent-skills

14

17

place-fact Route guidance, state, or evidence already worth keeping to one canonical owner and the narrowest reliable delivery b…

preetamnath/agent-skills

13

18

tighten-instruction Make an instruction line read cold: clarify it into plain words, then tighten it to one positive line. TRIGGER when: …

preetamnath/agent-skills

13

19

code-review Structured code review with P0-P3 findings, confidence scores, and criteria-based analysis. Use for reviewing code ch…

preetamnath/agent-skills

12

20

generate-mockups Generate UI mockups as one-file HTML, no build, grounded in the project's real design language — PREVIEW a feature's …

preetamnath/agent-skills

12

21

plan-builder Creates dependency-ordered, wave-grouped executable plans from a goal + context. Produces markdown plans with paralle…

preetamnath/agent-skills

11

22

multi-agent-analysis Dispatch 1–3 subagents to analyze or verify something, then judge their findings yourself. TRIGGER when: user says 'm…

preetamnath/agent-skills

10

23

test-completed-plan Run the live testing phase after execute-plan ships a build — drive the spec's '### Post-ship verification' checklist…

preetamnath/agent-skills

9

24

deep-modules The deep-module primitive: a small interface hiding lots of behaviour. Apply the lens to judge or shape one interface…

preetamnath/agent-skills

8

25

explain-deeply Explain a feature, flow, or concept: read the source of truth, lead with the answer, and diagram each idea the answer…

preetamnath/agent-skills

8

26

find-gaps Surface what a draft, answer, or code is missing — parallel subagents each apply a different lens, scoring findings b…

preetamnath/agent-skills

8

27

git-commit-message Generate git commit messages in a specific bullet-point format with a conventional header.

preetamnath/agent-skills

8

28

post-purchase-ui-extension SDK reference for `@shopify/post-purchase-ui-extensions-react` — Shopify's post-purchase upsell surface (the screen b…

preetamnath/agent-skills

8

29

second-opinion Second opinion on one concrete proposal: stress-test it, rank alternatives, or both. TRIGGER when: user says 'second …

preetamnath/agent-skills

8

30

fix-loop Bounded fix-and-verify protocol for resolving findings from a review. Max 2 attempts per finding, adaptive verificati…

preetamnath/agent-skills

7

31

polaris-web-components Polaris <s-*> web component catalog, props/slots reference, and layout patterns for the polaris-app-home surface. Use…

preetamnath/agent-skills

7

32

supervise-plan Keep a long execute-plan run alive across context windows: from a separate chat, watch the shared mailbox and respawn…

preetamnath/agent-skills

7

33

triage Verify a panel's findings against their artifact: each gets a consider / skip verdict and a confidence.

preetamnath/agent-skills

7

34

handoff Compress the live conversation into minimal, copy-paste-ready handoff prompt(s) — context referenced by file path, no…

preetamnath/agent-skills

6

35

map-design-language Research a project's design language and write or refresh a lean meta/DESIGN.md of facts — surfaces, toolkits, tokens…

preetamnath/agent-skills

6

36

post-purchase-extension Post-purchase UI extension SDK reference (29 React components, lifecycle, sandbox rules) for `@shopify/post-purchase-…

preetamnath/agent-skills

6

37

propose-alternatives Propose 2-3 genuinely different approaches to a problem with concrete trade-offs. Use when evaluating design choices,…

preetamnath/agent-skills

6

38

trim-spec Trim a spec, ADR set, or design doc down to what a builder needs, without losing a fact. TRIGGER when: user says a sp…

preetamnath/agent-skills

6

39

validate-answer Trust-check an answer, plan, or decision you already have — multiple identical independent reads converge per questio…

preetamnath/agent-skills

6

40

audit-transcripts-for-learnings Audit past Claude Code transcripts in a chosen scope and date window to extract reusable patterns, then walk each one…

preetamnath/agent-skills

5

41

compress-file Compress one instruction file to the leanest STRUCTURE that still delivers its purpose — dissolve sections that resta…

preetamnath/agent-skills

5

42

design-critique Critique already-built UI from screenshots, agree on the most useful review directions, and return ranked fixes suppo…

preetamnath/agent-skills

5

43

explain-visually Explain a feature, flow, or concept by grounding in source-of-truth, leading with the answer, and diagramming each lo…

preetamnath/agent-skills

5

44

memory-prune Prune project memory files: keep memory-native records, route guidance, state, and evidence to canonical owners, then…

preetamnath/agent-skills

5

45

sanity-check Validate or challenge a plan, design, or decision. Confirms what's good, flags realistic concerns, and identifies bli…

preetamnath/agent-skills

5

46

vet-fact Judge whether a candidate current-guidance fact earns a durable instruction line: keep only what prevents a likely wr…

preetamnath/agent-skills

5

47

diagnose-review-fix-orchestrator End-to-end pipeline for uncertain bugs: diagnose root cause, review the fix, then verify and harden. Chains parallel-…

preetamnath/agent-skills

4

48

panel-review Multi-reviewer panel on N focused questions about a near-final artifact (plan, design, code, prose). R0 (you) plus tw…

preetamnath/agent-skills

4

49

parallel-diagnosis Parallel independent root-cause diagnosis for uncertain bugs. Two agents investigate independently, then converge on …

preetamnath/agent-skills

4

50

prove-behavior Choose the smallest valuable automated test evidence and prove it can detect the defect it guards. TRIGGER when: beha…

preetamnath/agent-skills

4