poteto-help
Answer questions about Pi pstack setup, workflow selection, and prompt wording without starting the work.
Skill by FireTheDevil
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/firethedevil-poteto-help
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: productivity, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 26, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/poteto-help
What's inside
- Skill name
poteto-help- Description
- Answer questions about Pi pstack setup, workflow selection, and prompt wording without starting the work. Type /skill:poteto-help with a question.
- License
- MIT
- Instructions
- 62 lines in SKILL.md · view on GitHub
Files (3)
- SKILL.md
- prompting.md · references/
- recipes.md · references/
SKILL.md (first lines)
# Poteto help Answer the help question, offer at most one ready-to-send prompt, and link the source file you read. Do not execute examples or start the workflow. If the message requests actual work instead of help, identify [poteto-mode](../poteto-mode/SKILL.md) as the route and let the user invoke it. This help turn remains question-only. Infer the need from the message and chat. If unclear, ask one multiple-choice question: setup, starting a task, picking a skill, fixing a run, or customizing the workflow. Answer only the selected need. Read the routed file before recommending it; that file overrides this map. Use local source links, not invented public URLs or Cursor documentation. ## Setup and runtime Read [setup-pstack](../setup-pstack/SKILL.md) and [Herdsman runtime](../poteto-mode/references/herdsman-runtime.md) for setup questions. On lead session startup, the extension reconciles global profiles, creating missing ones and applying safe package description and prompt updates. Installed configuration, including models, thinking, and tools, remains user-owned. Conflicts and package capability changes hold the affected profile. Restart Pi after reported profile changes and inspect the effective definitions. Same-name globals override project definitions. Installation alone does not synchronize profiles. `/setup-pstack [role]` previews an optional confirmed project copy; it is not a model-budget wizard. Models and tools belong to effective discovered definitions. Preserve Sol, Luna, and Astra policy; never invent model availability or substitute a fallback. When effective definitions or availability are unknown and matter to the answer, mention that uncertainty. At most once per chat, ask whether the user wants setup guidance now or later. Do not inspect agents as a progress poll or change configuration from help. For cost questions, explain the cost of independent exploration and review, not nonexistent budget menus. For verification questions, identify whether a matching project verification skill or actual app tool is available. If none is known, say so and route to [create-verification-skill](../create-verification-skill/SKILL.md). For a formal property, race, or invariant beyond test reach with no existing formal check, route to [create-formal-verification-skill](../create-formal-verification-skill/SKILL.md), explicitly invoked with `/skill:create-formal-verification-skill`. It produces one project-local formal check, not app-driving verification or a feature-map maintenance loop. If a formal check already exists and its covered code changed after its recorded sync commit, recommend `/skill:maintain-formal-verification-skill` and read the [maintenance skill](../maintain-formal-verification-skill/SKILL.md). Do not offer creation again. Missing sync history is a maintenance uncertainty, not evidence that the old pass proves current code. For a reader-visible checked chart, recommend `/skill:statechart` and read [Statechart](../statechart/SKILL.md). Explain its scratch, trusted-map, tool, and PNG approval gates. Help does not run checks, install tools, or mutate definitions. A green build does not prove behavior. ## Start and select `/poteto-mode` enables the style for the session; `/poteto-mode off` disables it. `/skill:poteto-mode` also enables it. `/pstack-status` takes no arguments and reports Poteto Mode and bundled skill visibility without changing session state. Hidden skills remain explicitly callable as `/skill:<name>`; no automatic discovery accuracy is promised. For prompt wording, read [prompting](references/prompting.md) and [recipes](references/recipes.md). | Need | Source | | --- | --- | | Rigorous implementation or workflow routing | [poteto-mode](../poteto-mode/SKILL.md) | | Mechanics, ownership, or placement | [how](../how/SKILL.md) | | Historical reasons or a number's origin | [why](../why/SKILL.md) | | Plain explanation | [teach](../teach/SKILL.md) | | Restate the goals and underlying problem without starting work | [in-your-own-words](../in-your-own-words/SKILL.md) | | Rewrite a prompt deeper, not bigger, without doing the task | [sharpen](../sharpen/SKILL.md) | | Recent cross-session context on a topic | [recall](../recall/SKILL.md) | | Current-chat outcomes, real visuals, and every self-contained human decision | [ketchup](../ketchup/SKILL.md) | | Breakage outside a diff | [blast-radius](../blast-radius/SKILL.md) | | Types and module design | [architect](../architect/SKILL.md) | | Same brief, competing candidates and synthesis | [arena](../arena/SKILL.md) | | Partitioned coverage or races | [swarm](../swarm/SKILL.md) | | Independent adversarial review | [interrogate](../interrogate/SKILL.md) | | Prose cleanup or document structure | [unslop](../unslop/SKILL.md), [technical-writing](../technical-writing/SKILL.md) | | Reusable app proof or drift maintenance | [create-verification-skill](../create-verification-skill/SKILL.md), [maintain-verification-skill](../maintain-verification-skill/SKILL.md) | | Formal properties beyond test reach, with no existing check | [create-formal-verification-skill](../create-formal-verification-skill/SKILL.md) | | Existing formal checks after covered-code changes | [maintain-formal-verification-skill](../maintain-formal-verification-skill/SKILL.md) | | Reader-visible statecharts, agreed-chart skeletons, or code-to-spec comparison | [statechart](../statechart/SKILL.md) | | Performance measurement rigor | [benchmark-checklist](../benchmark-checklist/SKILL.md) | | A bespoke rigorous run | [figure-it-out](../figure-it-out/SKILL.md) | | Decision trail | [show-me-your-work](../show-me-your-work/SKILL.md) | | Session lessons or repeated mistakes | [reflect](../reflect/SKILL.md), [correct](../correct/SKILL.md) | | Personal workflow | [automate-me](../automate-me/SKILL.md) | For another installed skill, read its frontmatter and body before routing. Principles are one-rule skills that shape decisions, not extra orchestration layers. Explicit invocation uses `/skill:principle-<name>`. ## Playbooks and close calls Playbooks are files selected inside poteto-mode, not standalone commands. Read the matching file before recommending it. - How explains mechanics; Why investigates reasons; Teach explains the evidence plainly. - Arena compares the same brief; Swarm partitions checks. Neither creates automatic worktrees or cloud isolation. One writer owns each boundary. External worktrees need authorization and separate lead sessions. - Recall reconstructs recent topic context; Ketchup summarizes this chat. A specific old branch or session routes to [Session pickup](../poteto-mode/playbooks/session-pickup.md). - For a plan-only deliverable, use [Multi-phase plan](../poteto-mode/playbooks/multi-phase-plan.md). For unsettled design, use [Prototype](../poteto-mode/playbooks/prototype.md) or Architect first. State a checkpoint when implementation needs approval. - PR status or babysitting routes to [Babysit](../poteto-mode/playbooks/babysit.md). Status does not authorize fixes or merge. Landing routes to [Shipping](../poteto-mode/playbooks/shipping.md) with explicit authority and independent verification. - One task to a predicate routes to [Autonomous run](../poteto-mode/playbooks/autonomous-run.md). A multi-day program routes to [Orchestrate](../poteto-mode/playbooks/orchestrate.md). Neither grants scheduling or broader authority. - Pausing routes to [Pause safely](../poteto-mode/playbooks/pause-safely.md). Skill edits use [Authoring a skill](../poteto-mode/playbooks/authoring-a-skill.md); evaluations require an authorized available runner. ## Troubleshooting and reply For drift, restate the goal or say "new task". For overlapping writers, stop and resolve ownership rather than assuming isolation. For unsupported tools or models, report the missing capability. For claimed success from compilation alone, ask for inspectable proof of the real behavior. Lead with the answer. Give at most one adapted prompt from recipes and the source link. Apply unslop to the reply without turning help into implementation.
From FireTheDevil/pi-pstack-herdsman, MIT.
More from FireTheDevil/pi-pstack-herdsman
- 0.601score
ketchup
FireTheDevil
Catch up on what changed in this chat in plain words and small evidence-backed visuals, with concise human decisions.
SkillAFreshness: A★ 3Claude Code +9 - 0.581score
interrogate
FireTheDevil
Use for "interrogate", "adversarial review", "multi-model review", "challenge this", "stress test this code", "find blind spots", or "tear this apart".
SkillAFreshness: A★ 3Claude Code +9 - 0.581score
how
FireTheDevil
Use for "how does X work", code walkthroughs before changing something, and placement / ownership / layering questions ("where should this live", "which…
SkillAFreshness: A★ 3Claude Code +9 - 0.579score
automate-me
FireTheDevil
Use for "automate me", "create/update/refresh my -mode skill", "turn/capture my preferences or working style into a skill", or wanting agents to follow how…
SkillAFreshness: A★ 3Claude Code +9 - 0.579score
benchmark-checklist
FireTheDevil
Vet a perf measurement (limiter, tuning, limits, errors, repeatability, relevance, and whether the work happened) before you report or act on it.
SkillAFreshness: A★ 3Claude Code +9 - 0.578score
no-comments
FireTheDevil
Spawn Comment Sicko, fix accepted findings, and offer encodings for claimed constraints.
SkillAFreshness: A★ 3Claude Code +9 - 0.576score
blast-radius
FireTheDevil
Find what a change could break somewhere else before it ships, beyond the diff, and prove the one fact it's safe because of by running real code instead of…
SkillAFreshness: A★ 3Claude Code +9 - 0.576score
figure-it-out
FireTheDevil
Design an auditable playbook when no narrower one fits: a large migration, an ambitious multi-part change, or work reviewed after stepping away.
SkillAFreshness: A★ 3Claude Code +9 - 0.568score
in-your-own-words
FireTheDevil
Restate the user's goals and the problem they are trying to solve.
SkillAFreshness: A★ 3Claude Code +9
Signals
- Stars
- 3
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- Oct 15, 2026 (14 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from FireTheDevil
- 0.601score
ketchup
FireTheDevil
Catch up on what changed in this chat in plain words and small evidence-backed visuals, with concise human decisions.
SkillAFreshness: A★ 3Claude Code +9 - 0.581score
interrogate
FireTheDevil
Use for "interrogate", "adversarial review", "multi-model review", "challenge this", "stress test this code", "find blind spots", or "tear this apart".
SkillAFreshness: A★ 3Claude Code +9 - 0.581score
how
FireTheDevil
Use for "how does X work", code walkthroughs before changing something, and placement / ownership / layering questions ("where should this live", "which…
SkillAFreshness: A★ 3Claude Code +9 - 0.579score
automate-me
FireTheDevil
Use for "automate me", "create/update/refresh my -mode skill", "turn/capture my preferences or working style into a skill", or wanting agents to follow how…
SkillAFreshness: A★ 3Claude Code +9 - 0.579score
benchmark-checklist
FireTheDevil
Vet a perf measurement (limiter, tuning, limits, errors, repeatability, relevance, and whether the work happened) before you report or act on it.
SkillAFreshness: A★ 3Claude Code +9 - 0.578score
no-comments
FireTheDevil
Spawn Comment Sicko, fix accepted findings, and offer encodings for claimed constraints.
SkillAFreshness: A★ 3Claude Code +9
Cite this page
poteto-help skill. Agent Mart. https://agentmart.dev/skills/firethedevil-poteto-help (last updated 2026-10-24)