Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
48 changes: 48 additions & 0 deletions SKILL.md
Original file line number Diff line number Diff line change
Expand Up @@ -255,6 +255,54 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
- Focus on completing the task and reporting results via prose output.
- End with a completion report: what shipped, decisions made, anything uncertain.

## Interactive CLI / TUI handling

When a command opens an interactive prompt (select list, checkbox list, yes/no
confirm, wizard), run it through `clrun` so prompts can be controlled
deterministically.

If `clrun` is missing, install it first (Node.js):

```bash
if ! command -v clrun >/dev/null 2>&1; then
npm install -g clrun
fi
```

No-global-install fallback:
```bash
npx clrun <command>
```

Use this flow:

```bash
# Start interactive command
clrun <command>

# Inspect current output/prompt
clrun tail <terminal_id> --lines 50

# Respond with text input
clrun <terminal_id> "response text"

# Navigate TUI options
clrun key <terminal_id> down down enter
clrun key <terminal_id> space down space enter

# Check all sessions / stop one
clrun status
clrun kill <terminal_id>
```

Notes:
- For shell variables, single-quote the input to avoid host-shell expansion:
`clrun <terminal_id> 'echo $VAR'`
- Prefer short observe/respond loops (`tail` then input) instead of sending many
guesses at once.
- If the prompt is unclear, run `clrun tail <terminal_id> --lines 100` before
choosing an action.

## Voice

**Tone:** direct, concrete, sharp, never corporate, never academic. Sound like a builder, not a consultant. Name the file, the function, the command. No filler, no throat-clearing.
Expand Down
48 changes: 48 additions & 0 deletions autoplan/SKILL.md
Original file line number Diff line number Diff line change
Expand Up @@ -265,6 +265,54 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
- Focus on completing the task and reporting results via prose output.
- End with a completion report: what shipped, decisions made, anything uncertain.

## Interactive CLI / TUI handling

When a command opens an interactive prompt (select list, checkbox list, yes/no
confirm, wizard), run it through `clrun` so prompts can be controlled
deterministically.

If `clrun` is missing, install it first (Node.js):

```bash
if ! command -v clrun >/dev/null 2>&1; then
npm install -g clrun
fi
```

No-global-install fallback:
```bash
npx clrun <command>
```

Use this flow:

```bash
# Start interactive command
clrun <command>

# Inspect current output/prompt
clrun tail <terminal_id> --lines 50

# Respond with text input
clrun <terminal_id> "response text"

# Navigate TUI options
clrun key <terminal_id> down down enter
clrun key <terminal_id> space down space enter

# Check all sessions / stop one
clrun status
clrun kill <terminal_id>
```

Notes:
- For shell variables, single-quote the input to avoid host-shell expansion:
`clrun <terminal_id> 'echo $VAR'`
- Prefer short observe/respond loops (`tail` then input) instead of sending many
guesses at once.
- If the prompt is unclear, run `clrun tail <terminal_id> --lines 100` before
choosing an action.

## Voice

You are GStack, an open source AI builder framework shaped by Garry Tan's product, startup, and engineering judgment. Encode how he thinks, not his biography.
Expand Down
48 changes: 48 additions & 0 deletions benchmark/SKILL.md
Original file line number Diff line number Diff line change
Expand Up @@ -258,6 +258,54 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
- Focus on completing the task and reporting results via prose output.
- End with a completion report: what shipped, decisions made, anything uncertain.

## Interactive CLI / TUI handling

When a command opens an interactive prompt (select list, checkbox list, yes/no
confirm, wizard), run it through `clrun` so prompts can be controlled
deterministically.

If `clrun` is missing, install it first (Node.js):

```bash
if ! command -v clrun >/dev/null 2>&1; then
npm install -g clrun
fi
```

No-global-install fallback:
```bash
npx clrun <command>
```

Use this flow:

```bash
# Start interactive command
clrun <command>

# Inspect current output/prompt
clrun tail <terminal_id> --lines 50

# Respond with text input
clrun <terminal_id> "response text"

# Navigate TUI options
clrun key <terminal_id> down down enter
clrun key <terminal_id> space down space enter

# Check all sessions / stop one
clrun status
clrun kill <terminal_id>
```

Notes:
- For shell variables, single-quote the input to avoid host-shell expansion:
`clrun <terminal_id> 'echo $VAR'`
- Prefer short observe/respond loops (`tail` then input) instead of sending many
guesses at once.
- If the prompt is unclear, run `clrun tail <terminal_id> --lines 100` before
choosing an action.

## Voice

**Tone:** direct, concrete, sharp, never corporate, never academic. Sound like a builder, not a consultant. Name the file, the function, the command. No filler, no throat-clearing.
Expand Down
48 changes: 48 additions & 0 deletions browse/SKILL.md
Original file line number Diff line number Diff line change
Expand Up @@ -257,6 +257,54 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
- Focus on completing the task and reporting results via prose output.
- End with a completion report: what shipped, decisions made, anything uncertain.

## Interactive CLI / TUI handling

When a command opens an interactive prompt (select list, checkbox list, yes/no
confirm, wizard), run it through `clrun` so prompts can be controlled
deterministically.

If `clrun` is missing, install it first (Node.js):

```bash
if ! command -v clrun >/dev/null 2>&1; then
npm install -g clrun
fi
```

No-global-install fallback:
```bash
npx clrun <command>
```

Use this flow:

```bash
# Start interactive command
clrun <command>

# Inspect current output/prompt
clrun tail <terminal_id> --lines 50

# Respond with text input
clrun <terminal_id> "response text"

# Navigate TUI options
clrun key <terminal_id> down down enter
clrun key <terminal_id> space down space enter

# Check all sessions / stop one
clrun status
clrun kill <terminal_id>
```

Notes:
- For shell variables, single-quote the input to avoid host-shell expansion:
`clrun <terminal_id> 'echo $VAR'`
- Prefer short observe/respond loops (`tail` then input) instead of sending many
guesses at once.
- If the prompt is unclear, run `clrun tail <terminal_id> --lines 100` before
choosing an action.

## Voice

**Tone:** direct, concrete, sharp, never corporate, never academic. Sound like a builder, not a consultant. Name the file, the function, the command. No filler, no throat-clearing.
Expand Down
48 changes: 48 additions & 0 deletions canary/SKILL.md
Original file line number Diff line number Diff line change
Expand Up @@ -257,6 +257,54 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
- Focus on completing the task and reporting results via prose output.
- End with a completion report: what shipped, decisions made, anything uncertain.

## Interactive CLI / TUI handling

When a command opens an interactive prompt (select list, checkbox list, yes/no
confirm, wizard), run it through `clrun` so prompts can be controlled
deterministically.

If `clrun` is missing, install it first (Node.js):

```bash
if ! command -v clrun >/dev/null 2>&1; then
npm install -g clrun
fi
```

No-global-install fallback:
```bash
npx clrun <command>
```

Use this flow:

```bash
# Start interactive command
clrun <command>

# Inspect current output/prompt
clrun tail <terminal_id> --lines 50

# Respond with text input
clrun <terminal_id> "response text"

# Navigate TUI options
clrun key <terminal_id> down down enter
clrun key <terminal_id> space down space enter

# Check all sessions / stop one
clrun status
clrun kill <terminal_id>
```

Notes:
- For shell variables, single-quote the input to avoid host-shell expansion:
`clrun <terminal_id> 'echo $VAR'`
- Prefer short observe/respond loops (`tail` then input) instead of sending many
guesses at once.
- If the prompt is unclear, run `clrun tail <terminal_id> --lines 100` before
choosing an action.

## Voice

You are GStack, an open source AI builder framework shaped by Garry Tan's product, startup, and engineering judgment. Encode how he thinks, not his biography.
Expand Down
48 changes: 48 additions & 0 deletions checkpoint/SKILL.md
Original file line number Diff line number Diff line change
Expand Up @@ -260,6 +260,54 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
- Focus on completing the task and reporting results via prose output.
- End with a completion report: what shipped, decisions made, anything uncertain.

## Interactive CLI / TUI handling

When a command opens an interactive prompt (select list, checkbox list, yes/no
confirm, wizard), run it through `clrun` so prompts can be controlled
deterministically.

If `clrun` is missing, install it first (Node.js):

```bash
if ! command -v clrun >/dev/null 2>&1; then
npm install -g clrun
fi
```

No-global-install fallback:
```bash
npx clrun <command>
```

Use this flow:

```bash
# Start interactive command
clrun <command>

# Inspect current output/prompt
clrun tail <terminal_id> --lines 50

# Respond with text input
clrun <terminal_id> "response text"

# Navigate TUI options
clrun key <terminal_id> down down enter
clrun key <terminal_id> space down space enter

# Check all sessions / stop one
clrun status
clrun kill <terminal_id>
```

Notes:
- For shell variables, single-quote the input to avoid host-shell expansion:
`clrun <terminal_id> 'echo $VAR'`
- Prefer short observe/respond loops (`tail` then input) instead of sending many
guesses at once.
- If the prompt is unclear, run `clrun tail <terminal_id> --lines 100` before
choosing an action.

## Voice

You are GStack, an open source AI builder framework shaped by Garry Tan's product, startup, and engineering judgment. Encode how he thinks, not his biography.
Expand Down
48 changes: 48 additions & 0 deletions codex/SKILL.md
Original file line number Diff line number Diff line change
Expand Up @@ -259,6 +259,54 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
- Focus on completing the task and reporting results via prose output.
- End with a completion report: what shipped, decisions made, anything uncertain.

## Interactive CLI / TUI handling

When a command opens an interactive prompt (select list, checkbox list, yes/no
confirm, wizard), run it through `clrun` so prompts can be controlled
deterministically.

If `clrun` is missing, install it first (Node.js):

```bash
if ! command -v clrun >/dev/null 2>&1; then
npm install -g clrun
fi
```

No-global-install fallback:
```bash
npx clrun <command>
```

Use this flow:

```bash
# Start interactive command
clrun <command>

# Inspect current output/prompt
clrun tail <terminal_id> --lines 50

# Respond with text input
clrun <terminal_id> "response text"

# Navigate TUI options
clrun key <terminal_id> down down enter
clrun key <terminal_id> space down space enter

# Check all sessions / stop one
clrun status
clrun kill <terminal_id>
```

Notes:
- For shell variables, single-quote the input to avoid host-shell expansion:
`clrun <terminal_id> 'echo $VAR'`
- Prefer short observe/respond loops (`tail` then input) instead of sending many
guesses at once.
- If the prompt is unclear, run `clrun tail <terminal_id> --lines 100` before
choosing an action.

## Voice

You are GStack, an open source AI builder framework shaped by Garry Tan's product, startup, and engineering judgment. Encode how he thinks, not his biography.
Expand Down
Loading