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/dkritarth/context-kernel/mcp-testergit clone --depth 1 https://github.com/dkritarth/context-kernelWhat 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 | $0.00068 | $0.00422 |
| Opus 5 | $0.00034 | $0.00211 |
| Sonnet 5 | $0.00014 | $0.00084 |
| Haiku 4.5 | $0.00007 | $0.00042 |
Grade A, and why
mcp-tester 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.
What it actually says
Role
Verify a deployed context-kernel Worker behaves correctly, end to end, without touching source or curated data.
Checks (report each pass/fail with the decisive line only)
- Health / reachability — the Worker responds at its URL.
- Auth deny — an unauthenticated MCP request, and a request with a wrong token, both return
401and perform no KV read. - Read path (READ_TOKEN) —
list_sectionsreturns the expected section names.get_contextwith no arg returns non-empty Markdown; with a validsectionreturns just that section.get_metareturns a version + content hash.
- Scope isolation — the READ_TOKEN cannot invoke
append_journal(expect a401/403). - Write path (WRITE_TOKEN) —
append_journalwrites a test entry,list_journalreads it back. Use an obviously-tagged test note (e.g. tagsmoke-test) so it is easy to purge. - Cleanup — remind the owner to remove the smoke-test journal entry.
Constraints
- Never use the owner's real tokens in plaintext in committed files or logs. Read them from the environment / secret store.
- Never post real personal data as test input. Use synthetic notes.
- If any check fails, quote the shortest decisive error line and stop; do not attempt fixes to server code — hand back to the owner or the main session.
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 · 37 lines · 68 tokens per session scan A a544f61dfd17
mcp-tester is an agent published in the GitHub repository dkritarth/context-kernel (1 stars, last pushed 1mo ago), licensed MIT. It adds 68 tokens to every session and 422 once invoked, about $0.0003 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
omp
Runtime.
agent-workflows
How Projektor fits the agentic dev-tools stack, and the end-to-end multi-agent loop: lifecycle, coordination, and project management.
autoevolve-optimizer
Autonomous optimization loop for config artifacts (detection-index, context-router) - mutate, score deterministically, keep only improvements. Two code-enforced safety gates wrap the loop.
health-monitor
Deep health analysis of Evolving Lite - sentinel history, hook performance, recommendations.
overview
Duvlify defines six agent surfaces and four tools once, then adapts them to MCP, plain HTTP and WebMCP so they always agree.
rest-endpoints
The small, stable slice of the REST API that guides depend on, alongside the primary MCP surface.