@complexthings/superpowers-agent 8.4.1 → 9.0.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (47) hide show
  1. package/.agents/skills/caveman/SKILL.md +49 -0
  2. package/.agents/skills/diagnose/SKILL.md +117 -0
  3. package/.agents/skills/diagnose/scripts/hitl-loop.template.sh +41 -0
  4. package/.agents/skills/grill-with-docs/ADR-FORMAT.md +47 -0
  5. package/.agents/skills/grill-with-docs/CONTEXT-FORMAT.md +63 -0
  6. package/.agents/skills/grill-with-docs/SKILL.md +90 -0
  7. package/.agents/skills/handoff/SKILL.md +15 -0
  8. package/.agents/skills/improve-codebase-architecture/DEEPENING.md +37 -0
  9. package/.agents/skills/improve-codebase-architecture/HTML-REPORT.md +123 -0
  10. package/.agents/skills/improve-codebase-architecture/INTERFACE-DESIGN.md +44 -0
  11. package/.agents/skills/improve-codebase-architecture/LANGUAGE.md +53 -0
  12. package/.agents/skills/improve-codebase-architecture/SKILL.md +81 -0
  13. package/.agents/skills/karpathy-guidelines/SKILL.md +75 -0
  14. package/.agents/skills/prototype/LOGIC.md +79 -0
  15. package/.agents/skills/prototype/SKILL.md +30 -0
  16. package/.agents/skills/prototype/UI.md +112 -0
  17. package/.agents/skills/review/SKILL.md +78 -0
  18. package/.agents/skills/setup-matt-pocock-skills/SKILL.md +125 -0
  19. package/.agents/skills/setup-matt-pocock-skills/domain.md +51 -0
  20. package/.agents/skills/setup-matt-pocock-skills/issue-tracker-github.md +22 -0
  21. package/.agents/skills/setup-matt-pocock-skills/issue-tracker-gitlab.md +23 -0
  22. package/.agents/skills/setup-matt-pocock-skills/issue-tracker-local.md +19 -0
  23. package/.agents/skills/setup-matt-pocock-skills/triage-labels.md +15 -0
  24. package/.agents/skills/tdd/SKILL.md +109 -0
  25. package/.agents/skills/tdd/deep-modules.md +33 -0
  26. package/.agents/skills/tdd/interface-design.md +31 -0
  27. package/.agents/skills/tdd/mocking.md +59 -0
  28. package/.agents/skills/tdd/refactoring.md +10 -0
  29. package/.agents/skills/tdd/tests.md +61 -0
  30. package/.agents/skills/to-issues/SKILL.md +83 -0
  31. package/.agents/skills/to-prd/SKILL.md +76 -0
  32. package/.agents/skills/triage/AGENT-BRIEF.md +168 -0
  33. package/.agents/skills/triage/OUT-OF-SCOPE.md +101 -0
  34. package/.agents/skills/triage/SKILL.md +103 -0
  35. package/.agents/skills/ubiquitous-language/SKILL.md +93 -0
  36. package/.agents/skills/zoom-out/SKILL.md +7 -0
  37. package/.agents/superpowers-agent +120 -140
  38. package/.github/copilot-instructions.md +3 -0
  39. package/.github/workflows/main.yaml +1 -1
  40. package/AGENTS.md +97 -0
  41. package/README.md +32 -94
  42. package/package.json +3 -5
  43. package/skills/collaboration/leveraging-cli-tools/SKILL.md +48 -101
  44. package/skills/meta/create-agents-md/SKILL.md +182 -0
  45. package/skills/meta/create-agents-md/skill.json +9 -0
  46. package/skills/meta/writing-prompts/SKILL.md +64 -287
  47. package/skills/meta/writing-prompts/references/platforms.md +114 -0
@@ -0,0 +1,114 @@
1
+ # Platform formats
2
+
3
+ Exact locations, file formats, frontmatter, and argument syntax for reusable commands on each platform. Read the section for your target platform when saving the file.
4
+
5
+ ## Contents
6
+ - The convergence on SKILL.md
7
+ - Claude Code (skills + legacy commands)
8
+ - GitHub Copilot (prompt files + custom instructions)
9
+ - Cursor (commands + rules + skills)
10
+ - Quick pick
11
+
12
+ ## The convergence on SKILL.md
13
+
14
+ All three platforms are moving the same direction: a reusable command is becoming a skill — a directory containing a `SKILL.md` file with YAML frontmatter. Claude Code states custom commands "have been merged into skills"; Cursor 2.4 promotes Skills as the successor to commands and ships a `/migrate-to-skills` flow; GitHub Copilot reads `AGENTS.md` and a skills-style layout. Legacy formats still work everywhere, but prefer skills for new work where the platform supports them.
15
+
16
+ A minimal skill, identical in shape across platforms:
17
+
18
+ ```
19
+ my-command/
20
+ └── SKILL.md
21
+ ```
22
+
23
+ ```markdown
24
+ ---
25
+ name: my-command
26
+ description: What it does and when to use it.
27
+ ---
28
+
29
+ # My Command
30
+
31
+ <the prompt body>
32
+ ```
33
+
34
+ ## Claude Code
35
+
36
+ **Skills (recommended).**
37
+ - Project: `.claude/skills/<name>/SKILL.md` (commit to git to share)
38
+ - Personal: `~/.claude/skills/<name>/SKILL.md` (all your projects)
39
+ - The command name comes from the **directory name** → `/my-command`. Frontmatter `name` only sets the display label.
40
+ - Precedence: enterprise > personal > project. If a skill and a legacy command share a name, the skill wins.
41
+
42
+ **Legacy commands (still supported).**
43
+ - `.claude/commands/<name>.md` or `~/.claude/commands/<name>.md` → `/name` (from the file name).
44
+ - Support the same frontmatter as skills.
45
+
46
+ **Frontmatter (all optional; `description` is the one that matters for triggering):**
47
+ `name`, `description`, `when_to_use`, `argument-hint` (e.g. `[issue-number]`), `arguments` (named positional args), `disable-model-invocation` (default `false` — set `true` to make it manual-only, i.e. a pure slash command the model won't auto-run), `user-invocable` (default `true`), `allowed-tools` / `disallowed-tools`, `model` (or `inherit`), `effort` (`low`/`medium`/`high`/`xhigh`/`max`), `context` (`fork`), `agent`, `hooks`, `paths` (glob auto-activation), `shell` (`bash`/`powershell`).
48
+
49
+ **Arguments (0-based — note this differs from the old `$1`-is-first convention):**
50
+ - `$ARGUMENTS` — all arguments as one string. If the file contains none, the args are appended as `ARGUMENTS: <value>`.
51
+ - `$ARGUMENTS[0]` — first argument; `$ARGUMENTS[1]` second, etc.
52
+ - `$0`, `$1` — shorthand; **`$0` is the first argument.**
53
+ - `$name` — a named argument declared in `arguments:` frontmatter, bound by position.
54
+ - Also available: `${CLAUDE_SESSION_ID}`, `${CLAUDE_EFFORT}`, `${CLAUDE_SKILL_DIR}`.
55
+
56
+ **Bash injection in the body:**
57
+ - Inline: `` !`<command>` `` (only at line start or after whitespace).
58
+ - Block: a fenced ` ```! ` code block.
59
+ - Runs as preprocessing before the model sees the body; requires a matching `allowed-tools` grant such as `Bash(gh *)`.
60
+
61
+ **Skill vs command vs subagent:**
62
+ - Reference-style content (knowledge the model pulls in when relevant) → a normal skill.
63
+ - Action you invoke explicitly → a skill with `disable-model-invocation: true`, or a legacy command.
64
+ - Delegated, isolated-context work → a subagent in `.claude/agents/<name>.md` (frontmatter `name`, `description`, `tools`, `model`), or a skill run with `context: fork` + `agent:`.
65
+
66
+ ## GitHub Copilot
67
+
68
+ **Prompt files** (VS Code, Visual Studio, JetBrains):
69
+ - Location: `.github/prompts/`
70
+ - Naming: `<name>.prompt.md` (e.g. `explain-code.prompt.md`)
71
+ - Frontmatter documented by GitHub: `agent:` (value `'agent'`) and `description:`. (`model` and `tools` exist in VS Code's own prompt-file format but are **not** part of GitHub's documented spec — don't rely on them for portability.)
72
+ - Invoke: type `/<name>` (no extension) in Copilot Chat → `/explain-code`.
73
+
74
+ ```markdown
75
+ ---
76
+ agent: 'agent'
77
+ description: 'Generate a clear code explanation with examples'
78
+ ---
79
+
80
+ Explain the selected code. Cover what it does, the non-obvious parts, and one example call.
81
+ ```
82
+
83
+ **Repository custom instructions** (always-on context, not invoked):
84
+ - Repo-wide: `.github/copilot-instructions.md` — plain Markdown, no frontmatter.
85
+ - Path-specific: `.github/instructions/<name>.instructions.md` — frontmatter `applyTo:` (glob, e.g. `applyTo: "app/models/**/*.rb"` or `applyTo: "**"`) and optional `excludeAgent:` (`"code-review"` / `"cloud-agent"`).
86
+ - Copilot code review reads only the first 4,000 characters of an instruction file (other features are unaffected). Keep them tight.
87
+ - `AGENTS.md`, `CLAUDE.md`, and `GEMINI.md` are also recognized.
88
+
89
+ **Variables documented by GitHub:** input variables only — `${input:name}` and `${input:name:placeholder}` (e.g. `${input:code:Paste your code here}`). The `${selection}`, `${file}`, and `${workspaceFolder}` tokens come from VS Code's editor, not GitHub's spec — fine in VS Code, not portable.
90
+
91
+ ## Cursor
92
+
93
+ **Commands** (lightly documented; Cursor now steers users toward Skills):
94
+ - Location: `.cursor/commands/<name>.md` (project). A user-level form exists but Cursor doesn't publish the exact global path — don't quote one.
95
+ - No documented frontmatter; commands are described simply as reusable prompts.
96
+ - No documented argument/parameter passing.
97
+ - Invoke: type `/` in the Agent input and pick the command.
98
+
99
+ **Skills (preferred going forward):** `SKILL.md`-based, same shape as the convergence example above. Cursor ships `/migrate-to-skills` to convert commands.
100
+
101
+ **Rules** (passive context, auto-injected — not invoked like commands):
102
+ - Location: `.cursor/rules/` as `.md` or `.mdc`. Use `.mdc` for frontmatter.
103
+ - Frontmatter: `description` (string), `globs` (file pattern), `alwaysApply` (boolean).
104
+ - Types: **Always** (`alwaysApply: true`), **Agent Requested** (description, no globs — pulled in when relevant), **Auto Attached** (matches `globs`), **Manual** (only via `@`-mention).
105
+
106
+ ## Quick pick
107
+
108
+ | You want… | Claude Code | GitHub Copilot | Cursor |
109
+ |---|---|---|---|
110
+ | Invoke a saved prompt by name | `.claude/skills/<name>/SKILL.md` → `/name` | `.github/prompts/<name>.prompt.md` → `/name` | `.cursor/commands/<name>.md` → `/name` |
111
+ | Always-on project context | `.claude/skills/` (auto-invoked) or `CLAUDE.md` | `.github/copilot-instructions.md` | `.cursor/rules/*.mdc` |
112
+ | Path-scoped context | `paths:` frontmatter | `.github/instructions/*.instructions.md` (`applyTo`) | `.cursor/rules/*.mdc` (`globs`) |
113
+
114
+ When in doubt, write a `SKILL.md` — it's the format all three are converging on.