pstack — skill map

pstack is a Cursor plugin of ~45 skills built around one idea: an orchestrator mode (/poteto-mode) reads your task, picks a playbook, and routes to specialized skills and 21 engineering principles as steps fire. This document explains what each skill does and how they fit together.

1The architecture at a glance

👑 poteto-modesticky orchestrator · 22 playbooks · indexes principles inline
▼ matches your request to a playbook, then routes ▼
configuration & meta layer
setup-pstackmodel→role config rule
automate-medrafts your own “-mode”
figure-it-outdesigns ad-hoc playbooks
│ invoked by playbook steps │
capability layers
how · why · recall · blast-radiusunderstanding
architect · tdd · ts-best-practicesbuilding
arena · swarm · interrogateparallelism & review
create / maintain-verification-skillproof
│ every output passes through │
prose & record layer
unslop · technical-writing · no-comments · browriting quality
show-me-your-work · reflectdecision trail & learning loop
│ ground every decision │
foundation: 21 principles (leaf skills)
core ×9
architecture ×6
verification ×3
delegation ×2
meta ×1
│ spawned on demand │
subagents & externals
poteto-agentruns the whole style end-to-end
Comment Sickoread-only reviewer (via /no-comments)
cursor-team-kitdeslop · control-cli · control-ui (external)

2The orchestration core

Three skills define how work gets shaped. Everything else is a step they call.

/poteto-mode orchestrator

The entry point and sticky mode. Opens a todo list whose first item is reading its inline principles index, matches your request to one of 22 playbooks (bug fix, perf, feature, shipping, autonomous run, orchestrate…), copies the steps verbatim, and invokes the other skills as those steps fire. Stays on across turns; opts out only when you say so.

/setup-pstack config

Detects which models you can access and writes an always-applied rule mapping roles (precise code → sol, mechanical code → grok, prose/judgment → fable…) to models. Every multi-model skill reads this rule and falls back to defaults if absent.

/automate-me meta

Mines your recent transcripts and drafts a personal <your-name>-mode skill from how you actually work, routing through pstack underneath. Uses Cursor's built-in create-skill plus unslop.

/figure-it-out meta

The escape hatch when no bundled playbook fits: designs a rigorous, auditable playbook for the task at hand, scales rigor to the stakes, runs a hypothesis loop.

3Understanding & investigation

/how understand

Walkthroughs of subsystem architecture and runtime flow; placement/ownership questions ("where should this live"). Evidence comes from the code itself.

/why understand

Design rationale and history. Discovers available MCPs at runtime and queries each evidence category in parallel: source control, issue tracker, long-form docs, chat, infra observability, error tracking, analytics.

/recall understand

Rebuilds recent working context from your own chat history plus the shared record and hands back a tight current-state brief. Typically run before starting or resuming a task — often the first thing poteto-mode does on session pickup.

/teach understand

A composite: runs how + why and weaves both findings into one plain explanation, built up diagram by diagram. Use when you want genuine understanding, not a summary.

/blast-radius understand

For small-looking changes: finds what else they could break, and proves the one fact you're safe because of by running real code — not by asserting it.

4Building: design, parallelism, adversarial review

/architect build

Before code crosses a function boundary: settle the caller's usage, types, and module shape first, then stay in the loop while implementation fills in. Prevents jumping to code that locks in the wrong shape.

/tdd build

Failing test first, then the fix — only when there's a cheap local test path. Pairs with the fix-root-causes principle (reproduce first).

/typescript-best-practices build

Grounds the type-system-discipline principle in concrete TS syntax. Auto-applies whenever a .ts/.tsx file is touched.

/arena parallel

N parallel attempts at the same task across different models, then pick a base and graft the strongest parts of the losers into it. Best-of-N with recombination.

/swarm parallel

N parallel workers across different slices — coverage matrices, races, gauntlets, exploration — drained into one aggregated report. Embodies guard-the-context-window: bulk stays in children, summaries come back.

/interrogate parallel

Multi-model adversarial review: several LLM reviewers try to break a diff from independent angles, including a strict code-quality lens. The last gate before shipping contested designs.

5Verification infrastructure

/create-verification-skill verify

If a project has no scripted way to prove app behavior, generates a project-local verify skill with a feature map — drives the app the way a user would, any language or platform.

/maintain-verification-skill verify

Keeps that verify skill honest over time: parallel source readers per feature plus one live pass driving every feature; at most one PR of proven corrections.

Why this pair matters

They industrialize the prove-it-works principle ("verify against the real artifact, not a proxy"). Playbooks like shipping and autonomous-run lean on the generated verify skill so agents can prove their own work without a human in the loop.

6Prose, review hygiene & memory

/unslop prose

Removes AI writing tells. Applies to every reply poteto-mode writes ("any prose surface"), and to code comments and docs.

/technical-writing prose

Layered doc standard (Diátaxis + Google developer style + STE + Global English) for docs, RFCs, readmes, PR descriptions, and commit messages.

/no-comments prose

Strips comments before review: spawns the Comment Sicko subagent (read-only), fixes accepted findings, offers encodings for claimed constraints. Runs before review; complements the external /deslop which runs before commit.

/bro prose

Restates the last message in plain human language, no jargon. A one-shot readability valve.

/show-me-your-work record

A TSV decision trail (what / why / evidence / result) for long-running or unattended work. Local by default; committed when a reviewer needs to trust the result.

/reflect learning loop

After a long task lands: spawns three parallel review subagents over the transcript, surfaces learnings, and routes each one to a concrete edit on an existing skill.

7The 21 principles — the foundation everything cites

One short leaf skill per principle. poteto-mode carries an inline index of them and requires each applied principle to trace to a concrete decision. Standalone files exist so other skills can reference principles by name.

GroupPrinciples
Core ×9laziness-protocol · foundational-thinking · redesign-from-first-principles · subtract-before-you-add · minimize-reader-load · outcome-oriented-execution · experience-first · exhaust-the-design-space · build-the-lever
Architecture ×6model-the-domain · boundary-discipline · type-system-discipline · make-operations-idempotent · migrate-callers-then-delete-legacy-apis · separate-before-serializing-shared-state
Verification ×3prove-it-works · fix-root-causes · sequence-verifiable-units
Delegation ×2guard-the-context-window · never-block-on-the-human
Meta ×1encode-lessons-in-structure
Cross-links worth knowing: typescript-best-practices implements type-system-discipline; tdd operationalizes fix-root-causes; swarm embodies guard-the-context-window; create-verification-skill industrializes prove-it-works; reflect enacts encode-lessons-in-structure; arena executes exhaust-the-design-space.

8Subagents, automations & external pieces

poteto-agent

Subagent (subagent_type: "poteto-agent") that runs poteto's style end-to-end. Reads poteto-mode in full — including the principles index — before doing any work. Same wrapper poteto-mode routes through.

Comment Sicko

Read-only comment-reviewer subagent. Almost always invoked indirectly via /no-comments, not directly.

benny automation pack

Dormant pack under automations/benny/: triages Slack issue reports, reproduces confirmed bugs, fixes with real UI evidence. Not registered as slash skills; enabled per-repo via FOR_AGENTS.md.

cursor-team-kit (external)

Ships /deslop (pre-commit) and control-cli / control-ui (shipping CLIs/UIs). Referenced by poteto-mode but installed separately.

Cursor built-ins

Plan mode replaces planning skills; create-skill is used by automate-me and authoring playbooks; the built-in babysit is deliberately superseded by poteto-mode's babysit playbook.

9How they work together — a typical bug-fix run

1
poteto-mode+ setup-pstack config
You type /poteto-mode scroll drifts every 750ms, repro first, fix, verify. Mode opens its todo list, reads the principles index, matches the bug fix playbook. Model roles resolve from the rule written by setup-pstack.
2
understandhow · why · recall · blast-radius
Walks the scroll subsystem (how), pulls history on prior drift fixes from MCPs (why), rebuilds context if resuming a session (recall), checks what a naive fix could break (blast-radius).
3
fixtdd · architect · principles
Writes the failing repro test first (tdd), settles the timer-lifecycle types before editing (architect + model-the-domain), traces the symptom to its root cause instead of nil-guarding (fix-root-causes), keeps the diff minimal (laziness-protocol). Bulk exploration goes to swarm; genuinely contested design choices go to arena.
4
verifyverify skill · prove-it-works
Runs the project's generated verification skill against the real app — evidence, not "it compiles" (prove-it-works). If the project has no verify skill, create-verification-skill made one.
5
attack & polishinterrogate · no-comments · unslop · technical-writing
interrogate sends the diff to multiple models to break it. no-comments strips comments via Comment Sicko before review. The final reply and PR description go through unslop and the technical-writing standard.
6
record & learnshow-me-your-work · reflect
Unattended? Decisions were logged to the TSV trail (show-me-your-work). Afterward, reflect mines the transcript and turns what was learned into edits on existing skills — making the next run better (encode-lessons-in-structure).

10Full skill reference

SkillDoesRelationships
ORCHESTRATION & META
poteto-modeSticky orchestrator; 22 playbooks; routes all other skillsHub — indexes all principles, invokes nearly every skill below
setup-pstackMaps model roles to your available models via a config ruleRead by every multi-model skill (arena, swarm, interrogate, reflect…)
automate-meDrafts a personal "-mode" skill from your transcriptsUses create-skill + unslop; keeps pstack as base
figure-it-outDesigns auditable ad-hoc playbooks when none fitsLogs via show-me-your-work; alternative to bundled playbooks
UNDERSTANDING
howSubsystem walkthroughs; placement/ownership questionsComposed by teach; cited by poteto-mode triggers
whyDesign rationale via parallel MCP evidence queriesComposed by teach
recallRebuilds recent context into a current-state briefSession pickup / resume entry point
teachWeaves how + why into one plain explanation with diagramsComposite of how + why
blast-radiusFinds what a change could break; proves safety by running codeBridges understanding → verification
BUILDING & PARALLELISM
architectSettles types/signatures/module shape before codingGate before implementation in feature/refactor playbooks
tddFailing test first, then fixOperationalizes fix-root-causes
typescript-best-practicesTS-specific rules when reading/editing .ts filesGrounds type-system-discipline in syntax
arenaN attempts at same task; pick base, graft best partsExecutes exhaust-the-design-space; contrast with swarm
swarmN workers on different slices; one aggregate reportEmbodies guard-the-context-window
interrogateMulti-model adversarial review of a diffFinal gate after arena/swarm produce
VERIFICATION
create-verification-skillGenerates project-local verify skill + feature mapIndustrializes prove-it-works; consumed by shipping/autonomous playbooks
maintain-verification-skillKeeps verify skill honest vs. the appMaintains create-verification-skill's artifact
PROSE, RECORD & LEARNING
unslopRemoves AI writing tells from any surfaceApplies to every poteto-mode reply; used by automate-me
technical-writingDiátaxis + Google + STE + Global English standardDocs, RFCs, PRs, commit messages
no-commentsComment-free diffs before reviewSpawns Comment Sicko; pairs with external deslop (pre-commit)
broRestates last message in plain languageStandalone readability valve
show-me-your-workTSV decision trail for unattended workConsumed by figure-it-out; written during autonomous runs
reflectMines transcript; routes learnings into skill editsEnacts encode-lessons-in-structure; closes the loop
PRINCIPLES (21 leaf skills)
principle-*One rule each: core ×9, architecture ×6, verification ×3, delegation ×2, meta ×1Indexed inline by poteto-mode; referenced by name from other skills