ISA
Owns the Ideal State Artifact β the universal primitive that holds the articulated ideal state of any thing (project, task, application, library, infrastructure, work session) as a hard-to-vary explanation. The ISA is a single document that articulates the ideal state, drives the build, verifies the build, and records the evolution of understanding. Five workflows: Scaffold (generate a fresh ISA from a prompt at a specified effort tier), Interview (adaptive question-and-answer to fill in or deepen sections), CheckCompleteness (score an existing ISA against the tier completeness gate and report gaps), Reconcile (deterministic merge of an ephemeral feature-file excerpt back into the master ISA, keyed on stable ISC IDs), Seed (bootstrap a draft project ISA from an existing repository's README, code structure, and recent commits). Examples directory contains canonical-isa.md (the showpiece reference, fully populated across all twelve sections), e1-minimal.md (90-second task β Goal + Criteria only), e3-project.md (mid-size project with eight sections), e5-comprehensive.md (full application with all twelve sections plus a populated changelog history). Twelve-section body order is locked: Problem, Vision, Out of Scope, Principles, Constraints, Goal, Criteria, Test Strategy, Features, Decisions, Changelog, Verification. The skill is invoked automatically by the Algorithm at OBSERVE for any non-trivial task and may also be invoked directly by the user to scaffold or audit an ISA outside an Algorithm run. USE WHEN: any prompt mentions ideal state, ISA, ISC, ideal state criteria, ideal state artifact, project specification, hill-climb on a task, articulating done, scaffolding an ISA, interviewing for an ISA, checking ISA completeness, reconciling an ephemeral feature file back to a master ISA, or seeding an ISA from an existing project. NOT FOR creating new skills (use CreateSkill), running the Algorithm itself (the Algorithm invokes this skill), generating non-ISA artifacts (this skill owns the ISA primitive only), or after-the-fact write-ups like postmortems, decision logs, or engineering journals (those are retrospective; the ISA is a commitment-time scaffold for ideal state).
What this skill does
## π¨ MANDATORY: Voice Notification (REQUIRED BEFORE ANY ACTION)
**You MUST send this notification BEFORE doing anything else when this skill is invoked.**
1. **Send voice notification**:
```bash
curl -s -X POST http://localhost:31337/notify \
-H "Content-Type: application/json" \
-d '{"message": "Running the WORKFLOWNAME workflow in the ISA skill"}' \
> /dev/null 2>&1 &
```
2. **Output text notification**:
```
Running the **WorkflowName** workflow in the **ISA** skill to ACTION...
```
**This is not optional. Execute this curl command immediately upon skill invocation.**
# ISA β Ideal State Artifact
The ISA is the single document that articulates "done" for any thing whose ideal state we are pursuing β a project, an application, a library, infrastructure, a work session, an art piece, a strategic decision. It serves five identities simultaneously: ideal state articulation, test harness, build verification, done condition, system of record.
This skill owns the canonical template, the workflows that generate and refine ISAs, and the example library.
---
## The Twelve-Section Body (locked v6.2.0)
Every ISA may have up to twelve body sections. The tier completeness gate decides which are required at which effort tier; sections never appear empty. **Order is fixed**.
| # | Section | Purpose | Written At |
|---|---------|---------|------------|
| 1 | `## Problem` | What is broken or missing right now that makes the ideal state worth pursuing | OBSERVE |
| 2 | `## Vision` | What euphoric surprise looks like β experiential intent, 1β5 sentences | OBSERVE |
| 3 | `## Out of Scope` | Anti-vision β what is *not* included in this ideal state, declared upfront in prose | OBSERVE |
| 4 | `## Principles` | Substrate-independent truths (Deutsch reach) the work must respect | OBSERVE |
| 5 | `## Constraints` | Immovable architectural mandates that bound the solution space | OBSERVE |
| 6 | `## Goal` | The hard-to-vary spine β 1β3 sentences naming verifiable done | OBSERVE |
| 7 | `## Criteria` | Atomic ISCs (Ideal State Criteria) β one binary tool probe each, including derived `Anti:` ISCs | OBSERVE β EXECUTE |
| 8 | `## Test Strategy` | Per-ISC verification approach β `isc | type | check | threshold | tool` | OBSERVE/PLAN |
| 9 | `## Features` | Work breakdown β `name | description | satisfies: [ISC-Nβ¦] | depends_on: [featureβ¦] | parallelizable: bool` | PLAN |
| 10 | `## Decisions` | Timestamped decision log including dead ends; `refined:` prefix for Goal/ISC restructures | any phase |
| 11 | `## Changelog` | Conjecture / refuted-by / learned / criterion-now entries β Deutsch error-correction trail | LEARN |
| 12 | `## Verification` | Evidence that each ISC passed β quoted command output, file content, screenshot path | VERIFY |
---
## Three-Guardrail Taxonomy (Principles vs Constraints vs Anti-criteria)
Adjacent concepts. Distinguished by **who they bind**.
| Guardrail | Binds | Tone | Example | Lives In |
|-----------|-------|------|---------|----------|
| **Principles** | The *thinking* | Aspirational, generalizable | "User-facing systems prioritize responsiveness." | `## Principles` |
| **Constraints** | The *solution space* | Immovable, non-negotiable | "We do not roll our own cryptography β OAuth via industry-standard libraries only." | `## Constraints` |
| **Out of Scope** | The *vision* | Declared, explicit, prose | "Mobile native apps are not part of v1." | `## Out of Scope` |
| **Anti-criteria** | The *test surface* | Granular, testable, yes/no | "Anti: /admin returns 200 in v1 build." | `## Criteria` (with `Anti:` prefix) |
The first three are author-stated (declarative). Anti-criteria are derived β they are how Out of Scope, Constraints, and Principles become probe-able.
---
## Tier Completeness Gate (HARD at all tiers)
Mirrors the v6.1.0 thinking-floor non-relaxability. Required sections per tier:
| Tier | Required Sections |
|------|-------------------|
| **E1** | Goal, Criteria |
| **E2** | Problem, Goal, Criteria, Test Strategy |
| **E3** | Problem, Vision, Out of Scope, Constraints, Goal, Criteria, Features, Test Strategy |
| **E4** | All twelve sections |
| **E5** | All twelve + active Interview workflow run before BUILD |
**Project ISA override:** any `<project>/ISA.md` requires E3+ structure regardless of the active task's tier. The project file is the long-lived source of truth; one transient E1 task must not downgrade it.
`CheckCompleteness` workflow enforces this gate. A miss blocks `phase: complete` until the missing sections are filled in.
---
## Workflow Routing
Match the verb in the request to a workflow. When ambiguous, default to Scaffold for new ISAs and CheckCompleteness for audits.
| Verb / Intent | Workflow | File |
|---------------|----------|------|
| "scaffold", "create", "generate", "new ISA from this prompt", "extract feature as ephemeral" | **Scaffold** | `Workflows/Scaffold.md` |
| "interview me", "fill in the ISA", "deepen", "ask me questions" | **Interview** | `Workflows/Interview.md` |
| "check", "audit", "score this ISA", "is it complete?" | **CheckCompleteness** | `Workflows/CheckCompleteness.md` |
| "reconcile", "merge feature file back", "ephemeral β master" | **Reconcile** | `Workflows/Reconcile.md` |
| "seed", "bootstrap from this repo", "draft an ISA from existing code" | **Seed** | `Workflows/Seed.md` |
| "append decision", "append changelog", "append verification", "record C/R/L entry" | **Append** | `Workflows/Append.md` |
**When executing a workflow, output this notification directly:**
```
Running the **WorkflowName** workflow in the **ISA** skill to ACTION...
```
---
## Gotchas
The highest-information-density part of this skill. Each entry captures a non-obvious failure mode that has bitten real ISA work.
- **ID-stability is the cornerstone of Reconcile β never re-number on edit.** When the Splitting Test produces a finer-grained version of `ISC-7`, preserve `ISC-7` as the parent and add `ISC-7.1`, `ISC-7.2`, etc. Even when an ISC is dropped, leave a tombstone (`- [ ] ISC-N: [DROPPED β see Decisions YYYY-MM-DD]`). Reconcile keys on stable IDs; renumbering breaks ephemeral feature-file merges silently and the failure mode looks like "the worker's checkmarks didn't land in master."
- **Ephemeral files are derived views, never sources of truth.** Scaffold's `--ephemeral` mode produces a slice of the master ISA at `MEMORY/WORK/{slug}/_ephemeral/<feature>.md`. Workers operate against that slice; Reconcile merges back. Hand-editing master content from an ephemeral file is policy-forbidden β the master is what persists; the ephemeral is what gets archived.
- **The Changelog format is non-negotiable.** Every entry needs all four pieces (`conjectured`, `refuted by`, `learned`, `criterion now`) in that order. Append refuses to write a partial C/R/L; if any of the four is missing, the entry is a Decision, not a Changelog. The format is what makes the Deutsch error-correction trail auditable across sessions.
- **Project ISAs upgrade tier to `max(declared, E3)` regardless of the active task's tier.** A `<project>/ISA.md` is the long-lived system of record for a thing with persistent identity. One transient E1 task on the project must NOT downgrade the structural minimum. CheckCompleteness applies this override automatically.
- **Empty sections never appear.** The twelve-section body is a *capacity*, not a *requirement* at every tier. Sections required-but-empty for the tier are populated; sections not required and not yet written are simply absent from the file. CheckCompleteness distinguishes `present` / `thin` / `missing` / `empty` and only `empty` is acceptable for `Verification` before VERIFY phase.
- **Anti-criteria are derived from Out of Scope plus regression-prevention concerns.** They are how the prose-guardrails (Out of Scope, Constraints, Principles) become probe-able. At least one is required at every tier; the absence of an anti-criterion at OBRelated in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts β single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score β₯ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.