Borrowing it
Nothing to install: this file belongs to fold-run/fold. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/fold-run/fold/main/.claude/skills/mcp-inspector/SKILL.mdgit clone --depth 1 https://github.com/fold-run/foldWrote 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.
[](https://agentmods.dev/skills/fold-run/fold/mcp-inspector)<a href="https://agentmods.dev/skills/fold-run/fold/mcp-inspector"><img src="https://agentmods.dev/badge/skills/fold-run/fold/mcp-inspector.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00071 | $0.01294 |
| Opus 5 | $0.00036 | $0.00647 |
| Sonnet 5 | $0.00014 | $0.00259 |
| Haiku 4.5 | $0.00007 | $0.00129 |
Grade A, and why
mcp-inspector 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 7d 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.
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 — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Inspecting fold as a real client
make conformance answers "does fold pass". The Inspector answers "what
does a client actually see", which is the question you have when a bug
report says a tool is missing and the tests are green.
Requires node/npx. Nothing to install:
npx @modelcontextprotocol/inspector --cli <url> --transport http --method tools/list
Stand fold up
make build
FOLD_CONFIG=./fold.config.example.json ./fold --port 8080 --log-level debug
fold binds 127.0.0.1 by default and serves MCP at /mcp, so the
Inspector target is http://127.0.0.1:8080/mcp. --validate first if the
config is hand-edited. For a federation you can point at without real
upstreams, scripts/conformance.sh already stands up the everything-server
on UPSTREAM_PORT (3901) behind fold on GATEWAY_PORT (3902) — read it
for the fixture wiring rather than rebuilding it.
Auth: when auth.mode is required, pass the bearer yourself —
--header "Authorization: Bearer $TOK". Exit code 3 means fold rejected
the token, 4 means it is not listening at all. Set auth.mode to
disabled in a scratch config when the thing under test is not auth.
The recipes that earn their keep
The invisibility diff — the single most useful thing here, and the thing the conformance suite only checks for its own fixture. Run the same method against fold and against the upstream directly, and diff:
npx @modelcontextprotocol/inspector --cli http://127.0.0.1:3902/mcp --transport http \
--method tools/list --format json | jq -S '.result' > /tmp/via-fold.json
npx @modelcontextprotocol/inspector --cli http://127.0.0.1:3901/mcp --transport http \
--method tools/list --format json | jq -S '.result' > /tmp/direct.json
diff /tmp/direct.json /tmp/via-fold.json
In passthrough (single upstream, no namespace) that diff must be
empty. Anything in it is an invisibility-rule violation. With a namespace
configured, the only differences allowed are {namespace}__{name} on tool
and prompt names, merged lists, and per-principal policy filtering —
descriptions, schemas, _meta, and resource URIs must survive untouched.
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.
- 7d ago First seen · 113 lines · 71 tokens per session scan A ca013d0ed21d
mcp-inspector is a skill published in the GitHub repository fold-run/fold (1 stars, last pushed today), licensed Apache-2.0. It adds 71 tokens to every session and 1,294 once invoked, about $0.0004 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 skills, from other repositories
pi-loop-forensics
Diagnose pi-go agent loops and degenerate turns — "agent loop aborted", runaway thinking with no tool calls, repeated phrases. Discriminates genuine model repetition collapse from a race, a tool-parse failure, or a too-low guard, and A/B replays a seed session across providers.
nightly-session-watch
Nightly sweep of the last 24h of pi-go sessions — anomalous runs, loop aborts, tool error rates, token waste, real prompt-token spend, and whether the observation and palace pipelines are still recording. Triages each finding to the specialist skill that diagnoses it. Use for an unattended daily health check, or on…
config-loader-helper
Diagnose configuration-related failures, enumerate required env vars, and guide safe local test setup (no secrets).
kessoku-di
Kessoku compile-time DI with parallel initialization for Go. Use when writing or debugging kessoku providers/injectors, enabling async dependencies, migrating from google/wire, or fixing go:generate/codegen issues in Go services.
fix-suggester
Diagnose failures and propose minimal, test-backed fixes with verification and rollback instructions.
explain-error
Explain an error message with optional context.