Dd Debugger

What it does

When your agent uses it

• Asked what values a function receives
• What parameters look like at runtime
• Capture live data from running services without redeploying

Example prompts

• “/dd-debugger”

Requirements

• Python 3
• A credential in DD_API_KEY
• A credential in DD_APP_KEY

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

1. 1If you don't know the service, ask the user before proceeding.
2. 2Verify the service using pup debugger context to list environments with active instances. If multiple environments exist, ask the user…
3. 3Find a method using the dd-symdb skill (pup symdb search --view probe-locations)
4. 4Place a probe with capture expressions for the values you need
5. 5Watch events with --fields for compact output
6. 6Delete the probe when done

What it can do on your machine

Read from SKILL.md and the folder at commit 6a3c662. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • brew
  • jq

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  Links to these hosts (documentation or services it may open):

  • docs.datadoghq.com

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • DD_API_KEY
  • DD_APP_KEY

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Dd Debugger loads about 2.7k tokens when it runs. Until then it costs about 67 tokens; SKILL.md has 776 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~67

When it runs · the whole SKILL.md, loaded when a task matches

~2.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from DataDog/pup at commit 6a3c662, republished under its Apache-2.0 licence (© DataDog). 776 words, ~2,657 tokens.

name

dd-debugger

description

Live Debugger - inspect runtime argument/variable values in production by placing log probes on methods. Use when asked what values a function receives, what parameters look like at runtime, or to capture live data from running services without redeploying.

metadata.version

1.1.0

metadata.author

datadog-labs

metadata.repository

https://github.com/datadog-labs/agent-skills

metadata.tags

datadog,debugger,live-debugger,probes,dd-debugger

metadata.alwaysApply

false

brew tap datadog-labs/pack && brew install pup

# OAuth2 (recommended)
pup auth login

# Or use API keys
export DD_API_KEY="key" DD_APP_KEY="key" DD_SITE="datadoghq.com"

# 0. List environments (if multiple, ask user which to use)
pup debugger context my-service --fields service,language,envs

# 1. Create a probe with capture expressions (recommended)
#    Use --fields id to get just the probe ID back
pup debugger probes create \
  --service my-service \
  --env production \
  --probe-location "com.example.MyController:handleRequest" \
  --capture "request.id" \
  --capture "request.headers" \
  --ttl 1h \
  --fields id

# 2. Stream events with compact output
pup debugger probes watch <PROBE_ID> --timeout 60 --limit 10 \
  --fields "message,captures,timestamp"

# 3. Clean up
pup debugger probes delete <PROBE_ID>

# Full JSON output (default)
pup debugger context my-service

# Compact: just the fields you need
pup debugger context my-service --fields service,language,envs

# Filter to a specific environment
pup debugger context my-service --env production --fields service,envs,repo

# All probes
pup debugger probes list

# Filter by service
pup debugger probes list --service my-service

pup debugger probes get <PROBE_ID>

pup debugger probes create \
  --service my-service \
  --env staging \
  --probe-location "com.example.MyClass:myMethod" \
  --capture "user.name" \
  --capture "order.items[0].price"

pup debugger probes create \
  --service my-service \
  --env staging \
  --probe-location "com.example.MyClass:myMethod(int, java.lang.String)" \
  --capture "user.name"

# Capture specific fields using dot notation
--capture "user.name"
--capture "request.headers"

# Access array elements
--capture "orders[0].total"
--capture "items[len(items)].name"

# Chain member access
--capture "response.body.data.id"

# Capture multiple specific values
pup debugger probes create \
  --service my-service --env prod \
  --probe-location "OrderService:processOrder" \
  --capture "order.id" \
  --capture "order.items[0].price" \
  --capture "customer.email"

# Full snapshot — captures all arguments, locals, return value
pup debugger probes create \
  --service my-service --env staging \
  --probe-location "com.example.MyClass:myMethod" \
  --capture

# Full snapshot + specific expressions
--capture --capture "user.name"

# Template with capture expressions
pup debugger probes create \
  --service my-service --env prod \
  --probe-location "MyClass:myMethod" \
  --capture "user.id" \
  --template "Processing request for user={user.id}, took {@duration}ms"

--template "Processing order={orderId} for user={userId}"
--template "handleRequest took {@duration}ms"

--condition "status == 'error'"
--condition "@duration > 100"

pup debugger probes delete <PROBE_ID>

pup debugger probes watch <PROBE_ID>

# Message + captures + timestamp (most common)
pup debugger probes watch <ID> --fields "message,captures,timestamp" --limit 5

# Template message only (one line per event)
pup debugger probes watch <ID> --fields "message" --limit 10

# Just captures (expression values or snapshot data)
pup debugger probes watch <ID> --fields "captures" --limit 1

# Extract a specific captured expression
pup debugger probes watch <ID> --limit 1 \
  | jq '.snapshot.captures.return.captureExpressions'

# Get captured argument values
pup debugger probes watch <ID> --limit 1 \
  | jq '.snapshot.captures.return.arguments'

.snapshot.captures.return.arguments          → method arguments
.snapshot.captures.return.locals             → local variables
.snapshot.captures.return.captureExpressions → capture expression values
.snapshot.probe.id                           → probe ID
.snapshot.probe.location                     → source location

© DataDog, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/dd-debugger of DataDog/pup.

Open the folder on GitHubat commit 6a3c662