obligation-team

obligation-team is a command for Claude Code from prebid/salesagent. It costs 46 tokens per session (3,943 once invoked), scanned A, original, Apache-2.0.

A command that creates behavioral tests for specified obligations by assigning each obligation to a separate agent. The agents research the code, write tests, run them, and fix failures before the results are combined.

In plain words
What is it for?
It is for generating and verifying Python tests for obligation IDs, either individually or by selecting a number of matching IDs.
Why use it?
It speeds up test creation for many obligations while keeping each agent's temporary work separate.

Command for Claude Code

Written for Claude Code: arguments in frontmatter.

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/prebid/salesagent/obligation-team
Clone the repo
git clone --depth 1 https://github.com/prebid/salesagent

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for obligation-team

README.md
[![agentmods](https://agentmods.dev/badge/commands/prebid/salesagent/obligation-team.svg)](https://agentmods.dev/commands/prebid/salesagent/obligation-team)
Your own site
<a href="https://agentmods.dev/commands/prebid/salesagent/obligation-team"><img src="https://agentmods.dev/badge/commands/prebid/salesagent/obligation-team.svg" alt="Measured on agentmods" height="20"></a>
Per session 46 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,943 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00046 $0.03943
Opus 5 $0.00023 $0.01972
Sonnet 5 $0.00009 $0.00789
Haiku 4.5 $0.00005 $0.00394

Measured 4d ago against content hash 387afabfac43, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

obligation-team scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

PASS: @patch("...requests.post") — the delivery function calls requests.post.
.claude/commands/obligation-team.md · 461 lines

How it starts

The opening of the file, as written. The whole thing — 461 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Team-Based Obligation Test Generation: $ARGUMENTS

Spawn one agent per obligation. Each agent researches production code, writes a test to a temp file, runs it, fixes it until green, and reports back verified working code. Leader consolidates into the final test file and commits.

Architecture

  • Agents: Research + write temp test file + run pytest + fix until green (parallel)
  • Team lead: Consolidates verified tests into final file, quality gates, commit

Agents are full team members with shell access. They each write to their own temp file (tests/unit/_tmp_test_{OID_SUFFIX}.py) so there are no conflicts. The leader appends verified code to the shared test file after all agents report.

Protocol

Step 0: Resolve obligation IDs

Parse $ARGUMENTS to extract obligation IDs.

If prefix mode (args look like UC-004 --count 10 — no trailing sequence number):

python3 -c "
import json
al = json.loads(open('tests/unit/obligation_coverage_allowlist.json').read())
prefix = '$PREFIX'
matches = sorted(oid for oid in al if oid.startswith(prefix))
count = $COUNT  # default 10 if --count not specified
selected = matches[:count]
print('Selected obligations:')
for oid in selected:
    print(f'  {oid}')
print(f'Total matching: {len(matches)}, selected: {len(selected)}')
"

If direct IDs (args are full obligation IDs separated by spaces): Verify each ID exists in docs/test-obligations/ or the allowlist.

Store the resolved list as OBLIGATION_IDS.

Step 1: Gather shared context

Read these once — they'll be included in every agent's prompt:

  1. Test file header (imports + helpers):

    # Identify the test file from the use case prefix
    # UC-004 → tests/unit/test_delivery_behavioral.py
    head -110 tests/unit/test_delivery_behavioral.py
    
  2. Obligation scenarios for all resolved IDs:

    for OID in $OBLIGATION_IDS; do
      grep -A 30 "$OID" docs/test-obligations/*.md
    done
    
  3. Production files to read (list for the agent prompt):

    • src/core/tools/media_buy_delivery.py
    • src/core/schemas/delivery.py
    • src/core/webhook_delivery.py
    • src/core/webhook_authenticator.py

Read the full file on GitHub · 461 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 461 lines · 46 tokens per session scan A 387afabfac43

Subscribe to this mod's changes

obligation-team is a command published in the GitHub repository prebid/salesagent (37 stars, last pushed yesterday), licensed Apache-2.0. It adds 46 tokens to every session and 3,943 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-01.