Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

super-collect — find problems, hand them to the planning chain · erictechpro | DecimalAI

Productivityby erictechpro

super-collect — find problems, hand them to the planning chain

Find problems in one project and file them as tickets into its super-board Backlog — the board fixes them. Source plug-ins - sentry (app errors), posthog (exceptions, rage/dead clicks, failure events, web vitals, tracking gaps), github (issues not on the board), prs (recurring problems in merged PRs and their review comments), architecture (refactor findings), plus any custom source added in onboard (an MCP server, HTTP API, CLI command or app). Each candidate is checked by one fresh verifier be

Never made a case worseMaintained

SkillScore

SkillScore81/100

Test results 81

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

24%81%of test cases passed +57 pts

Without skill5 of 21

With skill17 of 21

12 fixed5 passed either way4 still failing0 broke1 errored

Model 11/3/2026v222 attempted

Track record

Agent runs: 0; All time

See one task, with and without the skill

One test taskWe are directly filing a verified Sentry issue with ID 98234 via the collection filing script. What fingerprint value should be passed to --fingerprint?

Without the skillMissed

With the skillPassed

Same model, same task. Answers are capped and redacted.

Use this skill

Your agent runs on

Adds every file the skill has, pinned so you can update it. No account needed.

pip install "decimalai>=0.15"
decimalai skills add erictechpro-super-collect
claude --plugin-dir "$(git rev-parse --show-toplevel 2>/dev/null || pwd)/.agents"

The same install from Python, for a notebook or a setup script.

Install

pip install "decimalai>=0.15"

Add the skill

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

print("Added. Start Claude Code with the command below.")

Start Claude Code with it

claude --plugin-dir "$(git rev-parse --show-toplevel 2>/dev/null || pwd)/.agents"

Type these in a Claude Code session. No account or API key. It writes the skill and any files it bundles, and asks first when its safety rating isn't Passed.

Add the DecimalAI plugin, once

/plugin marketplace add decimal-labs/decimalai-claude-plugins
/plugin install decimalai@decimalai

Install this skill

/decimalai:install erictechpro-super-collect

The plugin needs python3 and curl on your machine.

No Python or account. Claude Code picks it up in its next session.

Save it into your project

curl -fsSL --create-dirs -o .claude/skills/erictechpro-super-collect/SKILL.md https://app.decimal.ai/s/erictechpro-super-collect/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/erictechpro-super-collect/SKILL.md

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Claude Code docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[claude-agent-sdk]"
decimalai skills add erictechpro-super-collect

Add it to your agent

import asyncio

from claude_agent_sdk import ClaudeAgentOptions
from decimalai.claude_agent_sdk import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

async def main():
    question = "Hello!"
    async with support.run(question) as run:  # one user message, one run
        async for message in run.query(options=ClaudeAgentOptions(max_turns=6)):
            print(message)

# In a notebook or async code: await main()
asyncio.run(main())

Prefer its own loader? ClaudeAgentOptions(plugins=[skills_plugin()], skills="all") reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

The same install from Python, for a notebook or a setup script. agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[claude-agent-sdk]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

import asyncio

from claude_agent_sdk import ClaudeAgentOptions
from decimalai.claude_agent_sdk import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

async def main():
    question = "Hello!"
    async with support.run(question) as run:  # one user message, one run
        async for message in run.query(options=ClaudeAgentOptions(max_turns=6)):
            print(message)

# In a notebook or async code: await main()
asyncio.run(main())

Prefer its own loader? ClaudeAgentOptions(plugins=[skills_plugin()], skills="all") reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from claude_agent_sdk import ClaudeAgentOptions, query

skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
options = ClaudeAgentOptions(system_prompt=f"You are a helpful agent.\n\n{skill}", max_turns=6)

async def main():
    async for message in query(prompt="Hello!", options=options):
        print(message)

# In a notebook or async code: await main()
asyncio.run(main())

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Claude Agent SDK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[openai-agents]"
decimalai skills add erictechpro-super-collect

Add it to your agent

from agents import Agent, Runner
from decimalai.openai_agents import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    name=support.name,
    instructions=support.instructions,  # your prompt, then the skill list
    tools=[*support.tools],             # your tools, then load_skill
)
# In a notebook or async code: result = await Runner.run(agent, "Hello!")
result = Runner.run_sync(agent, "Hello!")

Using SandboxAgent? capabilities=[Shell(), decimalai.openai_agents.sandbox_skills()] points its own loader at the same folder.

Install

pip install "decimalai[openai-agents]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

from agents import Agent, Runner
from decimalai.openai_agents import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    name=support.name,
    instructions=support.instructions,  # your prompt, then the skill list
    tools=[*support.tools],             # your tools, then load_skill
)
# In a notebook or async code: result = await Runner.run(agent, "Hello!")
result = Runner.run_sync(agent, "Hello!")

Using SandboxAgent? capabilities=[Shell(), decimalai.openai_agents.sandbox_skills()] points its own loader at the same folder.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
agent = Agent(name="support", instructions=f"You are a helpful agent.\n\n{skill}")
# In a notebook or async code: result = await Runner.run(agent, "Hello!")
result = Runner.run_sync(agent, "Hello!")

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

OpenAI Agents docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[openai]"
decimalai skills add erictechpro-super-collect

Add it to your agent

import json

from openai import OpenAI
from decimalai.openai import agent_config

client = OpenAI()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your tools, as Chat Completions takes them, and what answers each.
TOOLS = [{"type": "function", "function": {
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters": {"type": "object", "properties": {"order_id": {"type": "string"}},
                   "required": ["order_id"]},
}}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    # run.system is your prompt, then the skill list.
    messages = [{"role": "system", "content": run.system}, run.message]
    while True:
        reply = client.chat.completions.create(
            model="gpt-5.4-mini",
            messages=messages,
            tools=[*TOOLS, *run.tools],  # your tools, then load_skill
        ).choices[0].message
        messages.append(reply)
        if not reply.tool_calls:
            break
        for call in reply.tool_calls:
            # load_skill is answered by the run; your own tools by FUNCTIONS.
            function = FUNCTIONS.get(call.function.name)
            messages.append(run.tool_result(call) or {
                "role": "tool", "tool_call_id": call.id,
                "content": function(**json.loads(call.function.arguments)) if function
                else f"There is no tool named {call.function.name!r}.",
            })
print(reply.content)

# The Responses API takes run.response_tools instead of run.tools.

Install

pip install "decimalai[openai]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

import json

from openai import OpenAI
from decimalai.openai import agent_config

client = OpenAI()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your tools, as Chat Completions takes them, and what answers each.
TOOLS = [{"type": "function", "function": {
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters": {"type": "object", "properties": {"order_id": {"type": "string"}},
                   "required": ["order_id"]},
}}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    # run.system is your prompt, then the skill list.
    messages = [{"role": "system", "content": run.system}, run.message]
    while True:
        reply = client.chat.completions.create(
            model="gpt-5.4-mini",
            messages=messages,
            tools=[*TOOLS, *run.tools],  # your tools, then load_skill
        ).choices[0].message
        messages.append(reply)
        if not reply.tool_calls:
            break
        for call in reply.tool_calls:
            # load_skill is answered by the run; your own tools by FUNCTIONS.
            function = FUNCTIONS.get(call.function.name)
            messages.append(run.tool_result(call) or {
                "role": "tool", "tool_call_id": call.id,
                "content": function(**json.loads(call.function.arguments)) if function
                else f"There is no tool named {call.function.name!r}.",
            })
print(reply.content)

# The Responses API takes run.response_tools instead of run.tools.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
reply = client.chat.completions.create(
    model="gpt-5.4-mini",
    messages=[
        {"role": "system", "content": f"You are a helpful agent.\n\n{skill}"},
        {"role": "user", "content": "Hello!"},
    ],
).choices[0].message

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

OpenAI SDK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[anthropic]"
decimalai skills add erictechpro-super-collect

Add it to your agent

import anthropic
from decimalai.anthropic import agent_config

client = anthropic.Anthropic()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

question = "Hello!"
with support.run(question) as run:  # one user message, one run
    answer = client.beta.messages.tool_runner(
        model="claude-haiku-5-5",
        max_tokens=4096,
        system=run.system,     # your prompt, then the skill list
        tools=[*run.tools],    # your tools, then load_skill
        messages=[{"role": "user", "content": question}],
    ).until_done()

Install

pip install "decimalai[anthropic]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

import anthropic
from decimalai.anthropic import agent_config

client = anthropic.Anthropic()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

question = "Hello!"
with support.run(question) as run:  # one user message, one run
    answer = client.beta.messages.tool_runner(
        model="claude-haiku-5-5",
        max_tokens=4096,
        system=run.system,     # your prompt, then the skill list
        tools=[*run.tools],    # your tools, then load_skill
        messages=[{"role": "user", "content": question}],
    ).until_done()

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
reply = client.messages.create(
    model="claude-haiku-5-5",
    max_tokens=4096,
    system=f"You are a helpful agent.\n\n{skill}",
    messages=[{"role": "user", "content": "Hello!"}],
)

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Anthropic docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[google-genai]"
decimalai skills add erictechpro-super-collect

Add it to your agent

from google import genai
from google.genai import types
from decimalai.google_genai import agent_config

client = genai.Client()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your functions, as declarations for the model, and what answers each.
TOOLS = [{"function_declarations": [{
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters_json_schema": {"type": "object", "properties": {"order_id": {"type": "string"}},
                               "required": ["order_id"]},
}]}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    config = types.GenerateContentConfig(
        system_instruction=run.system,     # your prompt, then the skill list
        tools=[*TOOLS, *run.tool_params],  # your functions, then load_skill
    )
    contents = [{"role": "user", "parts": [{"text": run.message}]}]
    while True:
        response = client.models.generate_content(
            model="gemini-3.5-flash", contents=contents, config=config,
        )
        if not response.function_calls:
            break
        contents.append(response.candidates[0].content)
        parts = []
        for call in response.function_calls:
            # load_skill is answered by the run; your own functions by FUNCTIONS.
            # A loop of your own rather than automatic function calling, which
            # raises KeyError when the model asks for a function you did not pass.
            function = FUNCTIONS.get(call.name)
            parts.append(run.tool_result(call) or {"function_response": {
                "name": call.name, "id": call.id,
                "response": {"result": function(**(call.args or {})) if function
                             else f"There is no function named {call.name!r}."},
            }})
        contents.append({"role": "user", "parts": parts})
print(response.text)

Install

pip install "decimalai[google-genai]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

from google import genai
from google.genai import types
from decimalai.google_genai import agent_config

client = genai.Client()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your functions, as declarations for the model, and what answers each.
TOOLS = [{"function_declarations": [{
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters_json_schema": {"type": "object", "properties": {"order_id": {"type": "string"}},
                               "required": ["order_id"]},
}]}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    config = types.GenerateContentConfig(
        system_instruction=run.system,     # your prompt, then the skill list
        tools=[*TOOLS, *run.tool_params],  # your functions, then load_skill
    )
    contents = [{"role": "user", "parts": [{"text": run.message}]}]
    while True:
        response = client.models.generate_content(
            model="gemini-3.5-flash", contents=contents, config=config,
        )
        if not response.function_calls:
            break
        contents.append(response.candidates[0].content)
        parts = []
        for call in response.function_calls:
            # load_skill is answered by the run; your own functions by FUNCTIONS.
            # A loop of your own rather than automatic function calling, which
            # raises KeyError when the model asks for a function you did not pass.
            function = FUNCTIONS.get(call.name)
            parts.append(run.tool_result(call) or {"function_response": {
                "name": call.name, "id": call.id,
                "response": {"result": function(**(call.args or {})) if function
                             else f"There is no function named {call.name!r}."},
            }})
        contents.append({"role": "user", "parts": parts})
print(response.text)

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

from pathlib import Path
from google import genai
from google.genai import types

client = genai.Client()
skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
response = client.models.generate_content(
    model="gemini-3.5-flash",
    contents="Hello!",
    config=types.GenerateContentConfig(system_instruction=f"You are a helpful agent.\n\n{skill}"),
)

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Gemini API docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[langchain]" langchain-openai
decimalai skills add erictechpro-super-collect

Add it to your agent

from langchain.agents import create_agent
from decimalai.langchain import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = create_agent(
    "openai:gpt-5.4-mini",
    tools=[],                         # your tools; the middleware adds load_skill
    middleware=[support.middleware],  # your prompt, the skill list and load_skill
)
result = agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

Install

pip install "decimalai[langchain]" langchain-openai

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

from langchain.agents import create_agent
from decimalai.langchain import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = create_agent(
    "openai:gpt-5.4-mini",
    tools=[],                         # your tools; the middleware adds load_skill
    middleware=[support.middleware],  # your prompt, the skill list and load_skill
)
result = agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
agent = create_agent("openai:gpt-5.4-mini", tools=[], system_prompt=f"You are a helpful agent.\n\n{skill}")
result = agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

LangChain docs

CrewAI's own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" "crewai>=1.15.27"
decimalai skills add erictechpro-super-collect

Add it to your agent

import decimalai.crewai
from crewai import Agent, Crew, Task

agent = Agent(
    role="Support", goal="Help the customer", backstory="A support agent.",
    llm=llm,                               # your own CrewAI LLM
    skills=decimalai.crewai.skills(),      # the installed folder
    tools=decimalai.crewai.skill_tools(),  # reads a skill's files, runs its scripts
)
task = Task(description="Hello!", expected_output="an answer", agent=agent)
Crew(agents=[agent], tasks=[task]).kickoff()

The same install from Python, for a notebook or a setup script. CrewAI's own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" "crewai>=1.15.27"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

import decimalai.crewai
from crewai import Agent, Crew, Task

agent = Agent(
    role="Support", goal="Help the customer", backstory="A support agent.",
    llm=llm,                               # your own CrewAI LLM
    skills=decimalai.crewai.skills(),      # the installed folder
    tools=decimalai.crewai.skill_tools(),  # reads a skill's files, runs its scripts
)
task = Task(description="Hello!", expected_output="an answer", agent=agent)
Crew(agents=[agent], tasks=[task]).kickoff()

No DecimalAI package. CrewAI's own loader reads the folder, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

from pathlib import Path
from crewai import Agent, Crew, Task

agent = Agent(
    role="Support", goal="Help the customer", backstory="A support agent.",
    llm=llm,                         # your own CrewAI LLM
    skills=[Path(".agents/skills")],
)
task = Task(description="Hello!", expected_output="an answer", agent=agent)
Crew(agents=[agent], tasks=[task]).kickoff()

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

CrewAI docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[adk]"
decimalai skills add erictechpro-super-collect

Add it to your agent

from decimalai.adk import DecimalAIPlugin, agent_config
from google.adk.agents import LlmAgent
from google.adk.apps import App

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = LlmAgent(
    name="support",          # ADK needs a Python identifier
    model="gemini-3.5-flash",
    instruction=support.instruction,  # your prompt, then the skill list
    tools=[*support.tools],           # your tools, then load_skill
)
# DecimalAIPlugin makes each invocation of runners built from this App one trace.
app = App(name="support", root_agent=agent, plugins=[DecimalAIPlugin()])

Prefer ADK's own loader? SkillToolset(skills=decimalai.adk.skills(), ...) reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

Install

pip install "decimalai[adk]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

from decimalai.adk import DecimalAIPlugin, agent_config
from google.adk.agents import LlmAgent
from google.adk.apps import App

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = LlmAgent(
    name="support",          # ADK needs a Python identifier
    model="gemini-3.5-flash",
    instruction=support.instruction,  # your prompt, then the skill list
    tools=[*support.tools],           # your tools, then load_skill
)
# DecimalAIPlugin makes each invocation of runners built from this App one trace.
app = App(name="support", root_agent=agent, plugins=[DecimalAIPlugin()])

Prefer ADK's own loader? SkillToolset(skills=decimalai.adk.skills(), ...) reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

from pathlib import Path
from google.adk.agents import LlmAgent

skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
agent = LlmAgent(
    name="support",
    model="gemini-3.5-flash",
    instruction=lambda context: f"You are a helpful agent.\n\n{skill}",
)

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Google ADK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[pydantic-ai]"
decimalai skills add erictechpro-super-collect

Add it to your agent

from pydantic_ai import Agent
from decimalai.pydantic_ai import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    "openai:gpt-4o-mini",
    name=support.name,
    tools=[],                 # your tools; the capability adds load_skill
    capabilities=[support],   # your prompt, the skill list and load_skill
)
# In a notebook or async code: result = await agent.run("Hello!")
result = agent.run_sync("Hello!")

On the Pydantic AI harness, decimalai.pydantic_ai_harness.skills(project) uses its own loader instead.

Install

pip install "decimalai[pydantic-ai]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

from pydantic_ai import Agent
from decimalai.pydantic_ai import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    "openai:gpt-4o-mini",
    name=support.name,
    tools=[],                 # your tools; the capability adds load_skill
    capabilities=[support],   # your prompt, the skill list and load_skill
)
# In a notebook or async code: result = await agent.run("Hello!")
result = agent.run_sync("Hello!")

On the Pydantic AI harness, decimalai.pydantic_ai_harness.skills(project) uses its own loader instead.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
agent = Agent("openai:gpt-4o-mini", instructions=f"You are a helpful agent.\n\n{skill}")
# In a notebook or async code: result = await agent.run("Hello!")
result = agent.run_sync("Hello!")

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Pydantic AI docs

Deep Agents' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" deepagents
decimalai skills add erictechpro-super-collect

Add it to your agent

import decimalai.deepagents
from deepagents import create_deep_agent

agent = create_deep_agent(model=model, **decimalai.deepagents.skills())
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

The default backend reads files but has no shell. To run a skill's scripts, pass skills(backend=LocalShellBackend(...)).

The same install from Python, for a notebook or a setup script. Deep Agents' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" deepagents

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

import decimalai.deepagents
from deepagents import create_deep_agent

agent = create_deep_agent(model=model, **decimalai.deepagents.skills())
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

The default backend reads files but has no shell. To run a skill's scripts, pass skills(backend=LocalShellBackend(...)).

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
agent = create_deep_agent(model=model, system_prompt=f"You are a helpful agent.\n\n{skill}")
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Deep Agents docs

Its own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" agent-framework
decimalai skills add erictechpro-super-collect

Add it to your agent

import asyncio

from agent_framework import Agent
from decimalai.agent_framework import skills

agent = Agent(client=client, instructions="You are a helpful agent.", context_providers=[skills()])
asyncio.run(agent.run("Hello!"))

It caches skills by default, so restart the process after decimalai skills update.

The same install from Python, for a notebook or a setup script. Its own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" agent-framework

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

import asyncio

from agent_framework import Agent
from decimalai.agent_framework import skills

agent = Agent(client=client, instructions="You are a helpful agent.", context_providers=[skills()])
asyncio.run(agent.run("Hello!"))

It caches skills by default, so restart the process after decimalai skills update.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
agent = Agent(client=client, instructions=f"You are a helpful agent.\n\n{skill}")
asyncio.run(agent.run("Hello!"))

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Microsoft Agent Framework docs

Strands' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" strands-agents
decimalai skills add erictechpro-super-collect

Add it to your agent

from decimalai.strands import skills
from strands import Agent
from strands.vended_tools import file_editor, shell

agent = Agent(
    model=model,                # always pass one: the default is Amazon Bedrock
    system_prompt="You are a helpful agent.",
    tools=[file_editor, shell],
    plugins=[skills()],
)
agent("Hello!")

Strands loads an agent's skills once, so build the agent again after decimalai skills update.

The same install from Python, for a notebook or a setup script. Strands' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" strands-agents

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("erictechpro-super-collect")
if result.exit_code:
    raise SystemExit(result.problems)

from decimalai.strands import skills
from strands import Agent
from strands.vended_tools import file_editor, shell

agent = Agent(
    model=model,                # always pass one: the default is Amazon Bedrock
    system_prompt="You are a helpful agent.",
    tools=[file_editor, shell],
    plugins=[skills()],
)
agent("Hello!")

Strands loads an agent's skills once, so build the agent again after decimalai skills update.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/erictechpro-super-collect/SKILL.md").read_text()
agent = Agent(model=model, system_prompt=f"You are a helpful agent.\n\n{skill}")
agent("Hello!")

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Strands docs

Adds every file the skill has, pinned so you can update it. Cursor, Copilot, Cline and Warp read this folder by themselves.

pip install "decimalai>=0.15"
decimalai skills add erictechpro-super-collect

No Python or account. Cursor, Copilot, Cline and Warp read this folder by themselves.

Save it into your project

curl -fsSL --create-dirs -o .agents/skills/erictechpro-super-collect/SKILL.md https://app.decimal.ai/s/erictechpro-super-collect/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/erictechpro-super-collect/SKILL.md

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Cursor and other editors docs

Fetch the file and give its text to your model as instructions. No account or package needed.

Get the file

curl -fsSL https://app.decimal.ai/s/erictechpro-super-collect/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/erictechpro-super-collect.json

This gets SKILL.md only. The skill's 9 other files are not included.

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v2, tested 11/3/2026
Source: erictechpro/super-board ↗
License: MIT

Manage it in DecimalAI

Saved prompts and skills, pinned versions, and traces of this skill in your runs. DecimalAI does not run your agent.

Claude Code reads files from your project, so DecimalAI config does not apply there.

On Claude Agent SDK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On OpenAI Agents: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On OpenAI SDK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Anthropic: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Gemini API: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On LangChain: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On CrewAI: set DECIMAL_API_KEY and add decimalai.init(crewai=True) to record runs and skill use. The skill version stays in your repo.

On Google ADK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Pydantic AI: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Deep Agents: set DECIMAL_API_KEY and add decimalai.init(langchain=True) to record runs and skill use. The skill version stays in your repo.

On Microsoft Agent Framework: set DECIMAL_API_KEY and add decimalai.init(otel=True, agent_name="support") to record runs and skill use. The skill version stays in your repo.

On Strands: set DECIMAL_API_KEY and add decimalai.init(otel=True, agent_name="support") to record runs and skill use. The skill version stays in your repo.

Editors read files from your project, so DecimalAI config does not apply there.

Anywhere else: record a use with decimalai.log_skill_activation() inside a trace.

Use this skill

SKILL.md

super-collect — find problems, hand them to the planning chain

One job: find problems and turn them into tickets. The flow:

Collect candidates from the sources.
One fresh verifier checks each candidate.
Write ONE findings doc that groups what survived: docs/super-board/findings/collect-<YYYY-MM-DD>.md.
/to-spec turns the findings doc into a spec.
/to-tickets turns the spec into tickets.
Build fork: 1–5 tickets → /implement-spec (one session, no QA lane). 6+ tickets → super-board: each ticket goes to Backlog, then /super-board lint and run.

No /to-spec or /to-tickets in this project → file each finding straight into Backlog (today's path, "Filing" below) and print why. super-collect never edits app code and never moves a card to Ready. Per project only: it reads the active project's super-board config and files onto that project's board. The whole rule in plain words: routing.

Invocation

Command	Does
`/super-collect`	every source enabled in `collect.sources`
`/super-collect <source> [<source>…]`	just those: `sentry`, `posthog`, `github`, `prs`, `architecture`, or a custom source's `name`
`--since 30d` / `--since 2026-09-01`	window; default `collect.window_days`, else 14 days
`--yes`	write and file without the confirm step (never skips verify or dedupe)
`--direct`	skip the planning chain: file each finding straight into Backlog

Default is dry-run: show the findings doc and the route, write nothing, go on only on confirm.

Scheduler integration

For periodic intake, use scheduled-intake.md: let the owner's scheduler invoke this existing skill, retain source verification and holding-column-only filing, and notify only on meaningful changes or failures. No job, cadence or scheduler is created by this skill.

Setup

Resolve the config: .claude/super-board/active → .claude/super-board/configs/<slug>.json (shape: ../super-board/references/config-schema.json). No config, or no collect block for a requested source → stop and point at /super-board onboard (its collect step sets sources up).
Secrets come only from .env (nearest walking up): SENTRY_AUTH_TOKEN, POSTHOG_PERSONAL_API_KEY. Never print, echo or paste them.
Source .claude/bin/super-board-gh-guard.sh; sb_gh_guard_check 200 before each gh burst.
Snapshot the board through .claude/bin/super-board-github-read.py --kind items -- project item-list <number> --owner <owner> --format json --limit 500. A count mismatch or read failure halts; never treat a partial board as complete. It is the dedupe set for verification and the "not on the board" set for github.

Scripts live in .claude/skills/super-collect/scripts/ (S= below). Every fetcher takes --since and prints JSON; a fetcher that cannot reach its source prints {"status":"unavailable","error":…} and exits 2. Record each source as read (n), empty, unavailable or truncated — never report an unreachable source as empty.

Sources

Source	Fetch	Candidate → ticket
`sentry`	`python3 $S/collect_sentry.py list --since <w>` — Sentry REST, Bearer token (scopes `event:read project:read`): org issues list, then `event <id>` / `tags <id>` for evidence	one unresolved issue → `bug`, fp `err
`posthog`	`python3 $S/collect_posthog.py list --since <w>` — HogQL `POST {host}/api/projects/{id}/query/`	one signal group → see table below
`github`	Read open candidates with `gh issue list --state open --json number,title,body,labels,url --limit 300` (a full 300 is capped coverage, not a complete repo inventory), minus issues on the board and `source:*` labels	real work → adopt (`--adopt <n>`), never copied
`prs`	`python3 $S/collect_prs.py list --since <w>` — one GraphQL search `is:pr is:merged merged:>=DATE`, paginated, with comments, reviews and review threads (human and bot), super-review reports flagged (`<!-- super-review:report -->`)	one recurring root cause → `fix` / `refactor` / `feature`, fp `prs
`architecture`	read-only finder sub-agent (below)	one finding → `refactor`, fp `arch
custom (`collect.custom[]`)	`python3 $S/collect_custom.py list --since <w>` — http and cli sources run in the script; an `mcp` source comes back `status: agent`: call that server's read/list tool yourself (read-only), then pipe its JSON into `collect_custom.py normalize --name <n>`	one candidate → `bug` (errors, incidents) or `feature` (requests), `--source custom`, fp `custom

custom sources

Added in onboard's "➕ Add another source" (super-board onboard → 📥 Bug sources): the user types a link, an app name, an API URL, an MCP server or a command; collect_custom.py classify decides which, ping proves it can be read (read-only) and add saves {name, kind, target, auth_env?, map?} to this project's config and its name to collect.sources. Here they run like any other source: every candidate goes to the same verifier and the same filer. A source that cannot be reached is unavailable, never empty. A command that looks like it changes something is refused by the script — never work around that.

posthog signals

The signal set is a table in collect_posthog.py (SIGNALS) plus the config's collect.posthog.failure_events; narrow it with collect.posthog.signals.

Signal	Grouped by	Files when	Ticket
`exception`	`$exception` by `issue_id` (follows merges)	≥ `min_users` (5) people	`bug`, label `error`
failure events	each configured `{event, fail}` (e.g. `push_completed` with `properties.status = 'failed'`)	≥ 5 people, or failing-people rate ≥ `failure_rate` (10%)	`bug`, label `error`
`rageclick`	`$rageclick` by URL + element	≥ 5 people	`bug`, label `ux`
`dead_click`	`$dead_click` by URL + element	≥ 5 people	`bug`, labels `ux`, `needs-triage` (high false-positive rate)
`web_vitals`	p75 per pathname	≥ 50 samples and p75 LCP > 4 s, INP > 500 ms or CLS > 0.25	`bug`, label `perf`
`survey`	raw `survey sent` responses (only when surveys exist)	you theme them; one theme from ≥ 3 people	`feature`

Each candidate carries a replay link (latest $session_id) — put it in Evidence. Funnel drop-off and abandonment spikes are report only: collect_posthog.py funnel --since <w> → <paths.runs_dir>/collect-funnel-<YYYY-MM-DD>.md. No tickets from them.

Tracking gaps (posthog runs only). collect_posthog.py events --since <w> lists every event seen plus silentSignals. Find the app's key user workflows from its routes and existing capture() calls; for each, list the success and failure events that are missing and the signals that return no data (e.g. autocapture off ⇒ no $rageclick). File one feature ticket per workflow with gaps, label analytics, fp gap|<workflow> — Problem, Context (routes, the capture calls found, what is missing), Fix, Acceptance Criteria (named events with their properties), Risk. Suggestions stay perf- and privacy-safe: sampling, clicks only, no input or text capture, no PII in properties. Never edit app code.

prs — what to look for

Read the JSON for problems that recur, not one-off nits:

The same finding shape (same module, same rule) on two or more PRs — from human reviewers, bots or super-review reports.
A super-review finding marked fixed that comes back not fixed in a later report.
Unresolved review threads on merged PRs that point at a real defect.
A skill, prompt or workflow whose reports keep missing the same thing → a feature to refine it.

Cluster by root cause at a shared boundary, not by wording. One ticket per cause with Problem, Context, Evidence (every PR, comment link and finding id), Fix (names the root cause), Acceptance Criteria, Risk. Before proposing Fix and its behavioral AC, follow correction prevention: identify what permitted the recurrence, prefer a fitting structural correction, and retain the actual prior bad witness.

architecture — the finder

Spawn one fresh, read-only sub-agent (it edits nothing and commits nothing). It uses, in order of what is installed: mattpocock improve-codebase-architecture (skip its HTML report and grilling; findings only) or codebase-design (module, interface, depth, seam, deletion test), plus ponytail:ponytail-audit for over-engineering. It reads CONTEXT.md and docs/adr/ and does not re-open a recorded decision; recently changed files weigh higher. It returns {module, problem, finding, files, why, severity}; each finding = one refactor ticket with Problem, Context (files, the shallow interface or duplicated concept), Fix, Acceptance Criteria (observable, plus "existing tests still pass"), Risk. Cap at collect.architecture.max_findings (5).

Verify — one fresh verifier per candidate

Before anything is filed, each candidate goes to one fresh verifier sub-agent. No for/against panel, no debate — one verifier, one verdict. Give it the candidate JSON, the board snapshot path and the window; nothing from other candidates. It checks, in order:

Real — the evidence holds up: open the Sentry event / replay / PR comment / code it cites.
Still happening — collect_sentry.py check <id> --after <window start> (seenAfter), or collect_posthog.py check <signal> <key> --after <date> (stopped, issueStatus).
Already fixed — a closed issue or merged PR that matches (gh pr list --state merged --search "<symptom> in:title,body", plus the fingerprint), and then from the fix's merge date or release: Sentry status: resolved with seenAfterRelease: false; PostHog stopped: true (zero hits and ≥ 3 days of traffic since the merge; with version_property, no hits on versions after the fix). A refactor already landed counts too.
Duplicate — same symptom + same route/module as a card in the snapshot or an open issue.

Verdict schema: {verdict: file | drop-fixed | drop-stale | drop-noise | duplicate | needs-triage, evidence: [links], note}. duplicate names the card; the filer then comments on it. Unclear is not a drop: needs-triage files the card with --label needs-triage. Report every drop with its evidence.

Hand-off — findings doc, /to-spec, /to-tickets

After Verify, pick the route:

python3 $S/collect_handoff.py route --root .     # {"route": "planning" | "backlog", "missing": [...], "reason": ...}

It resolves to-spec, to-tickets and implement-spec (Matt Pocock's skills, installed by get.sh) with the pack's skills-ready resolver. route: backlog (either of the first two missing) or --direct → skip to Filing below and print the reason line in the report.

route: planning:

Write the verifier's results as one JSON array (title, type, source, group, fingerprint, verdict, evidence, note). group is the shared root cause or area; findings with the same cause share one group.
python3 $S/collect_handoff.py doc --findings <verified.json> --since <w> prints the doc (dry run). On confirm, add --yes: it writes docs/super-board/findings/collect-<YYYY-MM-DD>.md (never overwrites; -2, -3, …) and prints the path. Only file and needs-triage findings go in; drops and duplicates are listed under "Not in this doc". A duplicate still gets its "Seen again" comment through the filer (--yes, same fingerprint), and a github candidate is still adopted with --adopt <n>: it is already a ticket.
Run /to-spec on the findings doc. It writes the spec; answer its seam question from the doc.
Run /to-tickets on that spec. Count the tickets it publishes.
python3 $S/collect_handoff.py fork --tickets <n> (add --no-implement-spec when route said implement_spec: null):
- 1–5 tickets → /implement-spec <spec>: one session, its own review, no QA lane. super-collect does not build: it prints the command; on confirm it starts it.
- 6+ tickets → super-board: add each ticket to Backlog with $S/super-collect-file.sh --config <cfg> --adopt <n> --type bug|feature|refactor --yes, then /super-board lint. Build → QA (with the security check) → Review → merge gate.
- Tickets in a local tracker (files, not GitHub issues) cannot go on the board: report their paths and stop.

With --yes and no one to answer /to-spec or /to-tickets (a scheduler), use --direct.

Filing

The direct path (no planning chain, --direct, adopting a github issue, a duplicate's "Seen again" comment) and the 6+ adoption go through one script, which routes to the pack's filers:

$S/super-collect-file.sh --config <cfg> --type bug|feature|refactor|fix \
  --source sentry|posthog|github|prs|architecture|custom --title "<one line>" --body-file <md> \
  --fingerprint "<key>" [--priority p] [--area a] [--label needs-triage] [--label ux] [--yes]
$S/super-collect-file.sh --config <cfg> --adopt <n> --type <t> [--yes]

bug, feature, fix → super-qa-file-bug.sh (fix files as tech-debt); refactor → super-review-file-refactor.sh. --title is the bare problem: the filers add the <emoji> [<kind>] <scope>: prefix, so never pass one (🐛 [bug] … would appear twice). Every body uses the ticket format in .claude/skills/super-board/references/writing-standard.md § 3: Problem · Context (lettered steps under Where) · Fix · Acceptance Criteria (checklist) · Risk; bugs add the folded 12-row Evidence table (a missing row says n/a — why), fixes add an Evidence section. Evidence is links and counts — never secrets or user PII. The filer refuses a body that misses a section.
Fingerprint per source (the filer rejects a mismatch): sentry err|sentry|<id>, posthog posthog|<signal>|<key> (fetchers emit it) or gap|<workflow>, github github|<n>, prs prs|<boundary>|<cause>, architecture arch|<module>|<problem>, custom custom|<name>|<key>.
Dedupe uses complete guarded paginated issue history, exact-fingerprint, repo-wide, any label: an open hit gets a "Seen again" comment; a closed-only hit is a recurrence, filed again naming the closed issue.
Column is the board's holding column (Backlog, else Todo/To do/Triage/Inbox); none → refused (exit 65).
The same-host repository lock serializes collector calls; digest receipts skip identical repeat comments. See state and recovery.
Labels: source:collect, collect:<source>, every --label, plus the filer's own.

Dry-run (direct path): run each verified item without --yes, show one table (would-file / duplicate / recurrence / would-adopt, plus the verifier's drops), wait for confirm, rerun with --yes.

Report

## super-collect: <planned n | filed n | dry-run n> · since 2026-09-18 (14d)
Coverage   sentry: read 42 | posthog: read 9 (silent: rageclick) | github: read 7 | prs: 23 PRs, 6 reports | architecture: 4 | linear: read 12
Findings   docs/super-board/findings/collect-2026-10-02.md  (7 findings, 3 groups)
Spec       #430  ← /to-spec
Tickets    #431–#438 (8)  ← /to-tickets
Build      6+ tickets → super-board: #431–#438 adopted into Backlog
Duplicate  #301 ← Sentry 4502 (comment added)
Dropped    Sentry 4490 drop-fixed — resolved in [email protected], not seen since (PR #377)
Report     docs/super-board/runs/collect-funnel-2026-10-02.md
Next       /super-board lint  — nothing here is in Ready

1–5 tickets: Build 3 tickets → /implement-spec #430 and Next /implement-spec #430. Direct path: a Route backlog — /to-tickets not installed in <root> … line, then today's Filed / Recurrence / Gaps lines (Filed #412 bug Checkout 500 on empty cart (err|sentry|4411)).

Common pitfalls

Filing one card per error event. Group by symptom and boundary first.
Skipping the verifier because the count looks big. Volume is not proof it is still happening.
Dropping an unclear candidate. File it needs-triage.
Copying a user's issue into a new one. Adopt it.
Treating a reviewer's fixed as proof. Recurrence is decided by the next report.
Editing app code, adding instrumentation, or moving anything to Ready. Collect plans and files; lint and run do the rest.
Writing one spec per finding. One findings doc per run; /to-spec makes one spec from it.

About this skill

Automatic

Project Management pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v8 — 11/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

erictechpro / super-board

skills/super-collect/SKILL.md

View on GitHub ↗

License MITImported 11/3/2026Version v2Bundled 9 files · scripts

Illustrative Want to see what erictechpro/super-collect would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Related skills

Skills that do similar work, with their headline benchmark lift.

From the same source and nearby in the catalogue.

Ready to supercharge your AI agents?

SkillScore

Safety

What it changes

Track record

Use this skill

super-collect — find problems, hand them to the planning chain

Invocation

Scheduler integration

Setup

Sources

custom sources

posthog signals

prs — what to look for

architecture — the finder

Verify — one fresh verifier per candidate

Hand-off — findings doc, /to-spec, /to-tickets

Filing

Report

Common pitfalls

SkillSafety

Source

Related skills

More in the registry