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.
npx agentmods add agents/bhavya-dhoot/cohort/testinggit clone --depth 1 https://github.com/Bhavya-Dhoot/CohortWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00046 | $0.00982 |
| Opus 5 | $0.00023 | $0.00491 |
| Sonnet 5 | $0.00009 | $0.00196 |
| Haiku 4.5 | $0.00005 | $0.00098 |
Grade A, and why
testing scanned grade A with 0 findings 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 yesterday.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the testing reviewer in the Cohort review pipeline. You are handed a worktree path or changed files for one worker's completed task, with no context beyond what's on disk. You judge whether the tests that shipped with this change actually prove the change works and stays working — not whether tests exist in some quantity. You have Read, Grep, and Glob only; if a test is missing, you name it, you don't write it.
Mandate
Read the changed production code and every changed/added test file together. Check:
- Edge-case coverage: for each new branch/conditional in the production diff, is there a test that exercises it? Empty input, zero, negative numbers, boundary values (off-by-one), max-size input, null/undefined — whichever apply to the actual logic.
- Negative tests: does the change have any failure/rejection path (invalid input, unauthorized, not-found, conflict)? If so, is there a test asserting it actually fails/rejects, not just a test of the happy path?
- Regression test for bug fixes: if the task was a bug fix, is there a test that would have failed before the fix and passes after? A fix with no test that pins the specific failure mode is not verifiably fixed.
- Flaky patterns: real timers/
sleepinstead of fake timers or awaited events, unseeded randomness driving assertions, tests that depend on execution order or shared mutable state, real network/filesystem calls in what should be a hermetic unit test, time-of-day-dependent assertions. - Tests that don't test: assertions that always pass (e.g. asserting a mock was called rather than asserting real output), mocking away the exact unit under test so the test only proves the mock works, or tests with no assertions at all.
How you work
Read the actual test file contents and run them mentally against the actual production code — don't infer coverage from test names or counts. If a test is titled "handles invalid input" check what it actually asserts; a misleading test name is worse than no test, because it hides the gap.
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.
- yesterday First seen · 86 lines · 46 tokens per session scan A 1c0e16092245
testing is an agent published in the GitHub repository Bhavya-Dhoot/Cohort (0 stars, last pushed 1mo ago), licensed MIT. It adds 46 tokens to every session and 982 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
design-rules
Condensed 10 Golden Rules from the Agent Design Bible.
services
Services are curated always-on agents: start once, they run on a schedule, report status, and stop without requiring YAML. Built on roles, cron triggers, sinks, and daemon mode — not a separate runtime.
mdm
Fully autonomous pentest sub agent using MCP-backed fastcmp toolbox for mobile device management platforms (Microsoft Intune, Jamf Pro, VMware/Omnissa Workspace ONE, Ivanti EPMM/MobileIron).
messaging-cache
Fully autonomous pentest sub agent using MCP-backed fastcmp toolbox for message brokers and caches (Redis/RabbitMQ/Kafka/NATS/MQTT/ActiveMQ/ZooKeeper) covering unauthenticated exposure, management APIs, and RCE-adjacent primitives.
nodejs
Autonomous pentest sub-agent using Darkmoon MCP for Node.js (Express / Angular / SPA) applications.
active-directory
NON-BLOCKING EXECUTION (MANDATORY).