MCPInstall

http-toolkit-intercept

Intercept and debug HTTP traffic from any CLI, service, or script using HTTP Toolkit. Use when you need to inspect LLM API calls, backend requests, auth flows, or debug network-level issues across any language or runtime.

Installs
118
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/factory-ai/factory-plugins --skill http-toolkit-intercept
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads http-toolkit-intercept automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

HTTP Toolkit Intercept

Use this skill when you need authoritative evidence of what your program sent to a remote API and what it received back while verifying a code change. Works across any runtime — Node.js, Bun, Deno, Python, Go, Ruby, Java/JVM, .NET, PHP, Rust, shell scripts, etc. — as long as the process respects a proxy.

The reliable pattern is:

  1. Start HTTP Toolkit correctly.
  2. Run the program through the proxy in a mode that produces a machine-readable log (e.g. --output-format json, debug logging, or structured stdout).
  3. Export outbound HTTP requests from HTTP Toolkit.
  4. Pair the outbound HTTP export with the inbound program session log.

Do not rely on TUI screenshots alone when the question is about request payloads, auth headers, or wire-level behavior.

Prerequisites / Known-Good Launch

Start HTTP Toolkit

On Linux/headless environments, plain httptoolkit often fails due to sandbox/X11 issues. Prefer:

xvfb-run --auto-servernum httptoolkit --no-sandbox

If a stale server is already running on ports 45456/45457, stop it first:

pkill -f "HTTP Toolkit Server|httptoolkit|xvfb-run --auto-servernum httptoolkit" || true

Verify the proxy is reachable

# HTTP Toolkit's admin API lives on port 45456/45457 by default.
# Treat 200/401/403 as "reachable"; only connection failure means the server is dead.
curl -s -o /dev/null -w "%{http_code}\n" http://127.0.0.1:45456/config

Quick Start

1. Launch your program with the proxy env vars set

The canonical env vars most runtimes honor:

HTTP_PROXY="http://127.0.0.1:8000" \
HTTPS_PROXY="http://127.0.0.1:8000" \
ALL_PROXY="http://127.0.0.1:8000" \
NO_PROXY="" \
<your-program> <args>

Some runtimes require an extra env var or flag — see the "Runtime proxy matrix" below.

If the HTTP Toolkit CA is not trusted by your runtime, TLS verification will fail. See "TLS Safety" below. Disabling TLS verification is appropriate only for controlled local debugging.

2. Capture the inbound session log

If your program supports a machine-readable output mode (e.g. --output-format stream-json, --json, --log-level debug, structured stdout, or writing to a file), pipe it to a file:

<your-program> exec --output-format stream-json "your input" \
  > /tmp/session-stdout.log 2> /tmp/session-stderr.log

3. Export outbound HTTP from HTTP Toolkit

Either:

  • Use the HTTP Toolkit GUI export (File → Export → JSON / HAR), or
  • Hit the admin API directly. The exact endpoint depends on your HTTP Toolkit version; inspect DevTools in the HTTP Toolkit UI to see the requests it makes.

4. Cross-reference the two streams

  • Outbound HTTP (from HTTP Toolkit): authoritative for request bodies, headers, auth tokens, retry timing.
  • Inbound session log (from the program): authoritative for how your code reacted to the responses.

Together they answer: "what did we send?" and "what did we do with the response?"

What finally worked for payload verification

The critical correct pathways that proved reliable were:

  1. Use a non-interactive / exec mode, not the TUI, when verifying payloads

    • Interactive TUIs are slower and much harder to analyze.
    • Use whatever your program has for scripting (--output-format, --json, --headless, --batch).
  2. Treat outbound and inbound as separate evidence sources

    • HTTP Toolkit gives outbound HTTP requests.
    • The program's session log gives inbound assistant/tool/application behavior.
    • You need both to answer: "what did the remote return?" and "what did the program actually do with it?"
  3. Set the proxy env vars your specific runtime honors

    • HTTP_PROXY/HTTPS_PROXY cover most runtimes, but some (e.g. Bun, Java) require extra vars or flags.
    • See the "Runtime proxy matrix" below.
  4. Disable TLS verification only for controlled local debugging when needed

    • If the HTTP Toolkit CA is not trusted locally, use the runtime-specific escape hatch to skip verification.
    • Prefer trusting the CA in your OS / language trust store instead.
    • Never disable TLS in production repros.
  5. Keep runs bounded

    • Long network-heavy sessions can take time.
    • If you only need to prove request shape, export after the relevant request is observed — you do not always need to wait for full completion.

Key Facts

  • Proxy env vars are runtime-specific — know which ones your runtime honors
  • HTTP Toolkit admin API is request-oriented — outbound HTTP comes from HTTP Toolkit, inbound behavior comes from the program log
  • TLS is verified by default — runtime-specific skip flags are local-dev escape hatches only

Runtime proxy matrix

RuntimeProxy env vars / flagsTLS bypass (local dev only)
Node.jsHTTP_PROXY, HTTPS_PROXY, NO_PROXY (most libraries); some HTTP clients need --proxy flags or explicit agent: optionNODE_TLS_REJECT_UNAUTHORIZED=0 or NODE_EXTRA_CA_CERTS=/path/to/ca.pem
BunBUN_CONFIG_PROXY (mandatory — plain HTTP_PROXY/HTTPS_PROXY are silently ignored for Bun's own fetch)NODE_TLS_REJECT_UNAUTHORIZED=0
DenoHTTP_PROXY, HTTPS_PROXY--unsafely-ignore-certificate-errors
Python (requests, httpx)HTTP_PROXY, HTTPS_PROXY, ALL_PROXYREQUESTS_CA_BUNDLE / SSL_CERT_FILE pointing at the HTTP Toolkit CA, or verify=False in code
Python (urllib)Same env varsSSL_CERT_FILE or disable context verification
Go (net/http)HTTP_PROXY, HTTPS_PROXY, NO_PROXY (honored via http.ProxyFromEnvironment)SSL_CERT_FILE or InsecureSkipVerify: true in the transport
RubyHTTP_PROXY, HTTPS_PROXYSSL_CERT_FILE, or OpenSSL::SSL::VERIFY_NONE
Java / JVM-Dhttp.proxyHost=127.0.0.1 -Dhttp.proxyPort=8000 -Dhttps.proxyHost=127.0.0.1 -Dhttps.proxyPort=8000; env vars are not honored by the JVMImport CA into a truststore and pass -Djavax.net.ssl.trustStore=...
.NET / C#HTTP_PROXY, HTTPS_PROXY, ALL_PROXY (on recent runtimes) or configure HttpClient explicitlyTrust CA in OS store, or HttpClientHandler.ServerCertificateCustomValidationCallback
PHP (curl / cli)HTTP_PROXY, HTTPS_PROXY (or per-call CURLOPT_PROXY)CURLOPT_SSL_VERIFYPEER => false
Rust (reqwest)HTTP_PROXY, HTTPS_PROXYdanger_accept_invalid_certs(true) on the client
curl / shellHTTP_PROXY, HTTPS_PROXY env vars or -x/--proxy flag-k / --insecure
Docker containersPass env vars through with -e HTTP_PROXY=...; use host.docker.internal (Mac/Win) or --network=host (Linux) so the container can reach the proxyMount the CA into the image and install it, or use runtime-specific bypass flags

Prefer trusting the HTTP Toolkit CA in your runtime/OS trust store over disabling verification. Bypass flags should be local-dev only.

TLS Safety Guardrails

  • Keep TLS verification enabled whenever possible.
  • Prefer trusting the HTTP Toolkit CA in your local trust store instead of disabling verification.
  • Use runtime-specific TLS-bypass flags only for controlled local debugging in development.
  • Never disable TLS when intercepting production traffic.

Inspecting Captured Logs

Filter to relevant events only

If your program emits newline-delimited JSON, use jq:

jq -c 'select(.type == "tool_call" or .type == "message")' /tmp/session-stdout.log

Adjust the filter to match your program's event schema. For plain-text logs, use rg/grep patterns.

Match outbound HTTP to inbound events

Sort both streams by timestamp, then interleave them. The sequence usually looks like:

outbound POST /api/endpoint        (from HTTP Toolkit)
inbound  event received             (from program log)
inbound  follow-up action           (from program log)
outbound POST /api/endpoint        (next request)

If a program log shows an outbound request that HTTP Toolkit didn't capture, that's a proxy-config bug.

Troubleshooting

ProblemCauseFix
Program hangs, no events after startupProxy env vars not reaching the process, or TLS verification blockingRe-run with the right env vars for your runtime (see matrix); if necessary, enable the runtime's TLS bypass in dev
ECONNRESET / connection reset on every requestRuntime-specific proxy env var missing (e.g. BUN_CONFIG_PROXY for Bun, JVM -D flags for Java)Use the correct runtime-specific proxy config
TLS cert errors via proxyMITM CA not trusted by this runtimeTrust HTTP Toolkit CA in the runtime / OS store, or enable the runtime's TLS bypass in dev only
HTTP Toolkit API 403s on /configAuth-gated config endpointTreat 200/401/403 as reachable; only connection failure means the server is dead
Export has outbound data but no matching inbound eventsDidn't capture the program logAdd > /tmp/session.log redirection to the launch
HTTP Toolkit misses the first requestStarted capturing after the process launchedStart HTTP Toolkit first, THEN launch the program
Container / VM can't reach 127.0.0.1:8000Loopback is container-localUse host.docker.internal (Docker Desktop) or --network=host (Linux)
Program ignores env vars entirelyRuntime doesn't honor env vars (e.g. JVM)Use runtime-specific flags (-Dhttp.proxyHost=... for JVM, etc.)

FAQ

What does the http-toolkit-intercept skill do?

Intercept and debug HTTP traffic from any CLI, service, or script using HTTP Toolkit. Use when you need to inspect LLM API calls, backend requests, auth flows, or debug network-level issues across any language or runtime.

How do I install http-toolkit-intercept?

Run npx skills add https://github.com/factory-ai/factory-plugins --skill http-toolkit-intercept in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/http-toolkit-intercept/ (personal) or .claude/skills/http-toolkit-intercept/ (project).

Which AI agents can use http-toolkit-intercept?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is http-toolkit-intercept safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains http-toolkit-intercept?

It is published by factory-ai in factory-ai/factory-plugins, and it is an official skill from the maker of the technology it covers. It has 118 recorded installs.

More from factory-ai/factory-plugins

View all →

no-use-effect

factory-ai/factory-plugins

Enforce the no-useEffect rule when writing or reviewing React code. ACTIVATE when writing React components, refactoring existing useEffect calls, reviewing PRs with useEffect, or when an agent adds useEffect "just in case." Provides the five replacement patterns and the useMountEffect escape hatch.

OfficialFrontend & Design1K installs

human-writing

factory-ai/factory-plugins

Remove signs of AI-generated writing from text to make it sound more natural and human-written. Use when editing or reviewing any form of document including: markdown, technical docs, emails, blog posts, PRDs, or any dedicated writing content. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.

OfficialDocs & Writing600 installs

simplify

factory-ai/factory-plugins

Open the skill to read its SKILL.md instructions.

Official240 installs

frontend-design

factory-ai/factory-plugins

Open the skill to read its SKILL.md instructions.

OfficialFrontend & Design164 installs

visual-design

factory-ai/factory-plugins

Image generation and presentations. Use when: - User asks for images: logos, icons, app assets, diagrams, flowcharts, architecture diagrams, patterns, textures, photo edits, restorations - User needs a presentation or slide deck Covers nanobanana CLI for image generation and Slidev for presentations.

OfficialFrontend & Design160 installs

security-review

factory-ai/factory-plugins

Open the skill to read its SKILL.md instructions.

OfficialCode Quality & Workflow157 installs

Popular backend & apis skills

View category →

find-skills

vercel-labs/skills

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

OfficialBackend & APIs4.1M installs

entra-app-registration

microsoft/azure-skills

Guides Microsoft Entra ID app registration, OAuth 2.0 authentication, and MSAL integration. USE FOR: create app registration, register Azure AD app, configure OAuth, set up authentication, add API permissions, generate service principal, MSAL example, console app auth, Entra ID setup, Azure AD authentication. DO NOT USE FOR: Key Vault secrets (use azure-keyvault-expiration-audit), general Azure resource security guidance.

OfficialBackend & APIs566.7K installs

This skill should be used when the user wants to "write agent code", "build an agent with ADK", "add a tool", "create a callback", "define an agent", "use state management" — in a project that needs ADK (Agent Development Kit) API patterns and code examples. It provides a quick reference for agent types, tool definitions, orchestration patterns, callbacks, state management, the graph Workflow API, and reference recipes to study. Do NOT use for scaffolding (use google-agents-cli-scaffold) or deployment (use google-agents-cli-deploy).

Backend & APIs514.8K installs

embedded-captions

heygen-com/hyperframes

Add captions or subtitles to an existing single-subject talking-head video without editing the footage. Use for plain verbatim captions, cinematic captions embedded behind the subject, VFX captions, “炸/特效/酷炫字幕,” or a named identity from the 35-style catalog. Route by visual identity, not by backend engine. The quiet anchor rail is the default; embed every word only when the user explicitly wants a fully cinematic treatment. The workflow runs locally end to end, including transcription and subject matting; split multi-shot footage before applying it.

Backend & APIs453.3K installs

entra-agent-id

microsoft/azure-skills

Provision Microsoft Entra Agent Identity Blueprints, BlueprintPrincipals, and per-instance Agent Identities via Microsoft Graph, and configure OAuth 2.0 token exchange (fmipath, OBO, cross-tenant) including the Microsoft Entra SDK for AgentID sidecar. USE FOR: Agent Identity Blueprint, BlueprintPrincipal, agent OAuth, fmipath token exchange, agent OBO, Workload Identity Federation for agents, polyglot agent auth, Microsoft.Identity.Web.AgentIdentities. DO NOT USE FOR: standard Entra app registration (use entra-app-registration), Microsoft Foundry agent authoring (use microsoft-foundry).

OfficialBackend & APIs336.6K installs

neon

neondatabase/agent-skills

Overview of Neon, a complete set of cloud backend primitives around Lakebase Postgres: Auth, Object Storage, Functions, the AI Gateway, and Realtime. Start here to choose Neon for undecided login, files, APIs, and LLM calls, set up the CLI or MCP server, and follow the branch-first workflow. Use when building an app or backend on Neon, or when "Neon" or "Lakebase Postgres" is mentioned. Child skill neon-postgres wins for an existing DATABASEURL, SQL, schema, inspect, or search. Child skill neon-auth wins for login, users, sessions, identity routing, and Managed Better Auth setup. Child skill neon-realtime wins for live queries in reactive apps and agents. Also use for object storage, S3, buckets, serverless functions, function triggers, cron, AI gateway, LLM calls, realtime, logs, Loki, Grafana, observability, postgres, database, backend, Claimable Neon, neon.new, or a no-signup database.

Backend & APIs237.1K installs