generate-appworld-code
Generate Python code to solve AppWorld agent tasks using playbook bullet guidance. Use when the AppWorld executor needs executable Python code for tasks involving Spotify, Venmo, Gmail, Calendar, Contacts, or other AppWorld APIs.
What this skill does
# Generate AppWorld Code
Generate executable Python code for AppWorld agent tasks, applying learned strategies from the ACE playbook.
## Purpose
When the AppWorld executor encounters a task, it calls this Skill with:
- Task instruction (natural language)
- Available apps (e.g., ['spotify', 'venmo'])
- Playbook bullets (learned strategies to apply)
You generate Python code that:
1. Solves the task using AppWorld APIs
2. Applies bullet guidance strategies
3. Handles errors gracefully
4. Calls `apis.supervisor.complete_task()` when done
## Input Format
```json
{
"instruction": "What is the title of the most-liked song in my Spotify playlists",
"apps": ["spotify"],
"strategies": [
"Always login before API calls",
"Handle pagination for large result sets"
],
"bullets": [
{
"id": "bullet-xxx",
"title": "Spotify login pattern",
"content": "Login to Spotify using apis.spotify.login() with credentials..."
}
]
}
```
## AppWorld API Patterns
### Spotify
```python
# Login
response = apis.spotify.login(username="[email protected]", password="password")
token = response["access_token"]
# Get playlists
playlists = apis.spotify.show_playlist_library(access_token=token)
# Get songs in playlist
songs = apis.spotify.show_playlist_songs(
access_token=token,
playlist_id=playlists[0]["id"]
)
```
### Venmo
```python
# Login
response = apis.venmo.login(username="[email protected]", password="password")
token = response["access_token"]
# Get friends
friends = apis.venmo.show_friends(access_token=token)
# Send payment
apis.venmo.send_payment(
access_token=token,
recipient_id=friend["id"],
amount=10.00,
note="Payment note"
)
```
### Gmail
```python
# Login
response = apis.gmail.login(username="[email protected]", password="password")
token = response["access_token"]
# Fetch emails
emails = apis.gmail.fetch_emails(
access_token=token,
max_results=10,
query="is:unread"
)
# Send email
apis.gmail.send_email(
access_token=token,
to="[email protected]",
subject="Subject",
body="Email body"
)
```
### Contacts
```python
# Get contacts
contacts = apis.contacts.show_contacts()
# Add contact
apis.contacts.add_contact(
name="John Doe",
email="[email protected]",
phone="+1234567890"
)
```
### Calendar
```python
# Get events
events = apis.calendar.show_events(
start_date="2025-01-01",
end_date="2025-12-31"
)
# Create event
apis.calendar.create_event(
title="Meeting",
start_time="2025-10-26T14:00:00",
end_time="2025-10-26T15:00:00"
)
```
## Code Generation Rules
1. **Always complete task**: Call `apis.supervisor.complete_task()` at the end
2. **Apply bullet strategies**: Use patterns from playbook bullets
3. **Handle errors**: Use try/except for API calls
4. **Be specific**: Don't use placeholders - generate actual implementations
5. **No explanations**: Return ONLY executable Python code
## Example Generation
**Input**:
```json
{
"instruction": "Send an email to [email protected] saying hello",
"apps": ["gmail"],
"strategies": ["Login before API calls", "Validate email addresses"],
"bullets": [...]
}
```
**Output**:
```python
# Gmail task: Send email to [email protected]
# Applying strategies: Login before API calls, Validate email addresses
try:
# Login to Gmail
response = apis.gmail.login(username="[email protected]", password="password")
token = response["access_token"]
# Validate recipient
recipient = "[email protected]"
if "@" not in recipient:
raise ValueError(f"Invalid email: {recipient}")
# Send email
apis.gmail.send_email(
access_token=token,
to=recipient,
subject="Hello",
body="Hello from AppWorld!"
)
# Complete task
apis.supervisor.complete_task()
except Exception as e:
print(f"Error: {str(e)}")
raise
```
## Credentials
AppWorld provides test credentials automatically. Use these common patterns:
- `username="[email protected]"`
- `password="password"`
- Tokens are returned from login APIs
## Common Patterns from Playbook
### Pattern: Friend Management (Venmo/Contacts)
```python
# Get current friends
current_friends = apis.venmo.show_friends(access_token=token)
current_ids = {f["id"] for f in current_friends}
# Get target friends (from contacts)
target_contacts = apis.contacts.show_contacts()
target_ids = {c["id"] for c in target_contacts if c.get("venmo_id")}
# Add missing
for target_id in target_ids - current_ids:
apis.venmo.add_friend(access_token=token, user_id=target_id)
# Remove extra
for current_id in current_ids - target_ids:
apis.venmo.remove_friend(access_token=token, user_id=current_id)
```
### Pattern: Aggregation (Spotify/Media)
```python
# Get all playlists
playlists = apis.spotify.show_playlist_library(access_token=token)
all_songs = []
for playlist in playlists:
songs = apis.spotify.show_playlist_songs(
access_token=token,
playlist_id=playlist["id"]
)
all_songs.extend(songs)
# Find most-liked
most_liked = max(all_songs, key=lambda s: s.get("likes", 0))
result = most_liked["title"]
```
## Response Format
Return Python code as plain text (no markdown formatting, no explanations).
The code should be immediately executable in the AppWorld environment.
Related 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 — 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 ≥ 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.