Dibs MCP server
Commit Nov 12, 2026Allows AI agents to submit software ideas to open source repositories and retain contributor credit through GitHub issue attribution.
Also known as hivecommons/dibs, dibs
- Stars
- 0
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- 0.1.0
- Last commit
- Nov 12, 2026
- License
- Apache-2.0
What is Dibs MCP?
Dibs allows AI agents to submit software ideas to open source repositories and retain contributor credit through GitHub issue attribution.
Summary generated from the project's README
About Dibs MCP server
Dibs is a marketplace designed to connect people with software ideas to open source repositories. It serves ideators who have insights but lack the time or desire to write code, and repository maintainers who use Hive to manage coding agents. The platform uses a language model to match proposed ideas with repositories that have declared an appetite for specific topics. Once a match is accepted by both parties, the system provides a prefilled GitHub issue for the user to file under their own account, ensuring native attribution through an ideated label.
Developers can interact with Dibs through a web interface at dibs.hivecommons.dev or by connecting an AI agent via a streamable HTTP MCP endpoint. Using the MCP server requires a bearer token for authentication. The system supports various input methods, including plain language, markdown, uploaded documents, and even transcribed audio. It distinguishes itself by treating conceptualization as a first-class contribution, recording settled ideas on a public credit wall to provide machine-readable provenance for every proposal.
What you can do with it
- 1Submit ideas via AI agentsAn AI agent can use the MCP endpoint to create a private draft idea including a title and description.
- 2Claim credit for software conceptsUsers can ensure their intellectual contributions are recognized by filing issues that link back to their Dibs profile.
- 3Match ideas to suitable projectsA language model analyzes ideas and repository profiles to suggest the best project matches.
- 4Convert audio to software proposalsUsers can submit ideas by providing audio files that are processed via a speech-to-text gateway.
Written from the project's README. Every use case is checked against a quote from it.
How to install Dibs MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall Dibs MCP in Claude Code
$ claude mcp add --transport http dibs https://dibs.hivecommons.dev/mcpInstall Dibs MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"dibs": {
"url": "https://dibs.hivecommons.dev/mcp"
}
}
}Install Dibs MCP in VS Code
.vscode/mcp.json
{
"servers": {
"dibs": {
"type": "http",
"url": "https://dibs.hivecommons.dev/mcp"
}
}
}Install Dibs MCP in VS Code CLI
$ code --add-mcp '{"name":"dibs","type":"http","url":"https://dibs.hivecommons.dev/mcp"}'Install Dibs MCP in Codex
$ codex mcp add dibs --url https://dibs.hivecommons.dev/mcpInstall Dibs MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.dibs] url = "https://dibs.hivecommons.dev/mcp"
Install Dibs MCP in Gemini CLI
$ gemini mcp add -s user --transport http dibs https://dibs.hivecommons.dev/mcpInstall Dibs MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"dibs": {
"url": "https://dibs.hivecommons.dev/mcp",
"type": "http"
}
}
}Configuration
DIBS_TOKENHUB_URL
Automated checks
Ran Nov 14, 2026Full report →Not verified yet. A server needs a score of 80, no serious finding, and enough data for us to check. How checks work · What to fix for more points
- Source repository linked and reachablehivecommons/dibs is public and reachablegithub.com/hivecommons/dibs ↗Pass
- Package names this repository as its sourceNot checked
- Build provenance (Sigstore attestation)Not checked
- Open source license (OSI)Apache-2.0spdx.org/licenses/Apache-2.0.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-11-12github.com/hivecommons/dibs ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNot checked
- No critical advisory in the package or its dependenciesNot checked
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeinitialize answered with protocol 2025-06-18 in 268 msPass
README
View on GitHubDibs
A place to offer software ideas to projects and retain credit for them.
Your idea, your credit, their code.
The missing contributor layer for the AI-agent era.
Dibs is a two-sided marketplace that connects ideators — people who call dibs on ideas: great insight, but no time, tooling, or inclination to write code — with hive-managed repositories: projects that use Hive to run coding agents under maintainer-chosen rules.
- Ideators post ideas — public or private — in plain language / markdown.
- Repos opt in to receive ideas, declaring topics and appetite.
- A language model (LLM) matches ideas to repos that want them; both the idea contributor and the receiving project must accept the match.
- On mutual acceptance, Dibs hands the ideator a prefilled GitHub
issue — they file it themselves, under their own GitHub account, so the
issue is natively attributed to them (label
ideated). The receiving project can then have its agents work on it under its own review rules. In this default flow Dibs is just the matchmaker: it never files the issue for you.
Ideators are first-class contributors
Open source has always had a quiet contributor class with no outlet: researchers, operators, designers, power users — people who know exactly what a project needs but will never open a pull request. Open source has also always known ideas are contributions — the all-contributors spec has a 💡 "ideas" emoji, and the academic CRediT taxonomy credits "Conceptualization" as a first-class role. Dibs makes that credit real, machine-readable, and public.
Creators get credit. Agents do the work. Projects take the bow.
Every implemented idea traces back to its ideator: the GitHub issue credits
the idea's author, the resulting work carries that provenance, and the public
credit wall (/api/credits, "Credit wall" in the UI) lists every ideator
with a settled idea: a GitHub issue URL has been recorded. This credits the
proposal, not proof that its implementation shipped or its issue was closed.
Settled ideas appear regardless of their visibility setting: settling a private
idea makes its credit facts (author, title/TLDR, target, issue URL, and time)
public, but not its body or private detail endpoint.
Try Dibs in your browser
No local installation or Go toolchain is needed. Open dibs.hivecommons.dev to browse the public surface; posting needs a Hive Hub sign-in and filing the issue needs a GitHub account.
- Sign in and describe your idea in plain language. Choose public or private deliberately; unsettled private ideas are not included in the public wall.
- Review suggested repositories and offer the idea to a suitable project. For Hive-managed targets, both sides accept before issue filing.
- Open the prefilled GitHub issue, review it, and file it yourself.
- Confirm the resulting issue URL in Dibs. This step is settlement: it records the handoff, not completion of the proposed work. Confirming also publishes the credit facts above, even for an idea marked private.
Non-Hive GitHub targets can skip the receiving-project acceptance step. For settlement and privacy details, see the consumer API. To submit from a coding agent instead, see MCP setup.
Maintainer and developer reference
The following sections cover running Dibs, its API, and local development, not prerequisites for using the hosted site.
Architecture
Single Go binary (cmd/dibs) serving both the JSON API and an embedded
static UI, deployed at its own subdomain: dibs.hivecommons.dev. The base
path is configurable (DIBS_BASE_PATH), so path-prefixed reverse-proxy
deployments also work.
| Package | Purpose |
|---|---|
pkg/server |
HTTP server, base-path routing, embedded UI |
pkg/auth |
Auth bridge — validates the hive hub session cookie against the hub |
pkg/store |
Idea model + JSON file store (atomic writes, no external DB), offer/settlement state machine |
pkg/fsutil |
Shared atomic JSON persistence helper (temp file + rename, mode 0600) used by the file stores |
pkg/api |
Idea CRUD + matching/offer/feed/decide/notification API (author-scoped, private ideas never leak) |
pkg/registry |
Hive-managed repo profiles + "accepting ideas" opt-in |
pkg/match |
LLM idea↔repo scoring via litellm gateway, cached TLDRs, "✨ Embellish" draft refinement, deterministic keyword fallback |
pkg/settle |
Settlement — prefilled GitHub new-issue URL the ideator files themselves (ideated label, 🐝 Dibs footer, URL-length budget) + issue-URL confirmation; legacy token mode. Ideas can target ANY GitHub repo: non-hive targets skip acceptance and their issues carry a "request a hive" growth CTA |
pkg/notify |
In-app notification feed (bell): matches, offers, decisions, issues |
pkg/mcpserver |
Streamable HTTP MCP endpoint exposing Dibs tools (see Submit ideas from your agent) |
pkg/intake |
Extracts idea text from uploaded documents or transcribed audio |
pkg/game |
Gamification layer — points, levels, badges — recomputed from stored idea state, never a mutable counter |
pkg/catalog |
Builds and searches a separate CNCF project catalog |
pkg/history |
Backfills public GitHub activity for hive-managed repos; persists daily counts for repo index charts |
pkg/indexformula |
Weighted composite behind the repo market index |
pkg/news |
Turns recent merged PRs into repo news cards |
pkg/deps |
Single place declaring the subsystem dependencies the HTTP server, JSON API, and MCP endpoint are wired with |
Configuration
| Env var | Default | Meaning |
|---|---|---|
DIBS_BASE_PATH |
/ |
Base path every route/asset is served under (set e.g. /ideas for path-prefixed proxying) |
DIBS_ADDR |
:8080 |
Listen address |
DIBS_METRICS_ADDR |
(unset) | Internal Prometheus /metrics listener (e.g. :9090); unset keeps it off. See deploy/ |
HUB_URL |
https://hive.hivecommons.dev |
Hive hub for session validation & repo sync |
DATA_DIR |
/data |
JSON store directory |
REPOS_SEED_FILE |
(unset) | Static JSON seed of repo profiles for dev/demo |
DIBS_LLM_BASE_URL |
(unset) | OpenAI-compatible gateway (hive litellm), e.g. http://litellm:4000/v1. Unset → deterministic keyword matcher |
DIBS_LLM_API_KEY |
(unset) | Bearer token for the LLM gateway |
DIBS_LLM_MODEL |
gpt-4o-mini |
Model name routed by the gateway |
DIBS_GITHUB_TOKEN |
(unset) | Legacy mode only for issue creation: when set, Dibs opens the credited issue server-side on accept. Unset (default), the ideator files a prefilled GitHub issue themselves — native attribution. Also used, when present, to authenticate the GitHub lookup that verifies a confirmed issue URL (anonymous otherwise) |
DIBS_ADMINS |
(unset) | Comma-separated, case-insensitive GitHub logins granted admin access (idea-moderation endpoints) |
DIBS_CLANKER_MARKERS |
hive: agent=,hivecommons-hive[bot],kubestellar-hive[bot] |
Comma-separated, case-insensitive markers matched against PR branch/labels/body/author to classify agent activity |
DIBS_INTAKE_MAX_MB |
25 |
Max upload size (MB) accepted for idea-intake file uploads |
DIBS_INTAKE_EXTRACT_WORKER |
(on) | Set to off to extract uploaded documents in-process instead of in a memory-limited, time-limited child worker (not recommended: crafted documents can then exhaust the server's memory) |
DIBS_STT_URL |
(unset) | Speech-to-text gateway URL; when set, enables voice/audio idea submission |
DIBS_STT_KEY |
(unset) | API key for the STT gateway |
DIBS_STT_MODEL |
(unset) | Model name routed by the STT gateway |
Legacy IDEATE_* names (the pre-rename prefix) are still honored as
fallbacks for every DIBS_* variable.
Roadmap
See ROADMAP.md for what is planned next. Delivery history:
- Wave 1 (done): scaffold, hub auth bridge, idea CRUD, repo registry.
- Wave 2 (done): LLM matching + TLDRs, swipe UX ("offer" / "pass"), issue settlement (credited GitHub issues), notifications.
- Wave 3 (done): deployment (GHCR image +
deploy/manifests for dibs.hivecommons.dev), public landing page, public credit wall, ideator profile stats. - Wave 4 (done): matchmaker settlement — the ideator files a prefilled GitHub issue themselves (native attribution,
ideatedlabel) and confirms the issue URL; LLM-assisted idea refinement on posting and repo-tailored expansion on submission; token-based settlement demoted to legacy. - Later: cookie-domain/hub-OAuth session sharing, non-hive receptor repos, idea economics.
Deployment
Every push to main publishes ghcr.io/hivecommons/dibs (latest +
commit sha) via .github/workflows/docker.yml; the image is distroless and
runs as non-root. Plain Kubernetes manifests for dibs.hivecommons.dev
(Deployment, Service, cert-manager TLS Ingress, PVC, ConfigMap/Secret) live
in deploy/, including the required DNS record and
hub-cookie prerequisites.
Public surface (no sign-in required)
| Route | Purpose |
|---|---|
GET / |
Landing page (pitch + sign-in) — signed-in users get the full app |
GET /api/credits |
Credit wall data: settled ideas only (handle, title, TLDR, repo, issue URL) |
GET /api/leaderboard |
Aggregate ideator scores/levels/badges |
GET /api/ticker |
Public idea ticker tape for the logged-out landing page |
GET /api/board |
Live-markets board shown on the logged-out landing page |
GET /api/stats |
Aggregate, non-identifying counts |
GET /api/repos/{org}/{repo}/index |
Repo value-index chart series |
GET /api/repos/{org}/{repo}/news |
Repo news feed |
GET /api/repos/{org}/{repo}/qr.png |
Repo QR code |
GET /healthz |
Liveness check + embedded git version |
GET /readyz |
Readiness check — verifies the data store is accessible/writable |
Everything else requires a hive hub session; signed-in ideators additionally
get GET /api/me/stats (ideas posted / offered / accepted / settled).
For response schemas, eligibility/privacy rules, settlement actions, rewards, and stability levels, see the credit-wall consumer API.
Submit ideas from your agent
Dibs exposes a stateless Streamable HTTP MCP endpoint at
https://dibs.hivecommons.dev/mcp, so Claude Code, Copilot CLI, Cursor,
Gemini CLI, Codex, and other MCP clients can submit ideas without leaving the agent.
Dibs is listed in the official MCP Registry as dev.hivecommons/dibs
(registry entry).
Search for that full name: a plain dibs search also returns unrelated
servers with the same name, including the 1stDibs marketplace and other
coding-agent tools called dibs. How the listing is published is covered in
MCP Registry publishing.
Available tools:
| Tool | Auth | Purpose |
|---|---|---|
submit_idea(title, description, tags?) |
Required | Creates a private draft idea owned by you and returns its id and API URL |
list_my_ideas() |
Required | Lists your ideas with lifecycle phase/status |
get_idea(id) |
Optional for public ideas; required for private ideas (author, or owner of a repo holding a pending or accepted offer) | Shows detail, matches, offers, and settlement state |
list_repos() |
Not required | Lists registered repo venues |
Token
Use Authorization: Bearer <token>. Dibs first asks the Hive Hub
/api/saas/whoami endpoint to verify the bearer token server-to-server; the
hub already accepts GitHub bearer tokens for registered hub users. If that is
not practical, copy the raw hive_hub_user cookie value from your browser
after signing in at https://hive.hivecommons.dev; Dibs falls back to verifying
that value through the same hub whoami cookie flow used by the web UI.
In Chrome/Edge: open DevTools → Application → Cookies →
https://hive.hivecommons.dev (or https://dibs.hivecommons.dev) → copy the
hive_hub_user value.
Claude Code
claude mcp add --transport http dibs https://dibs.hivecommons.dev/mcp \
--header "Authorization: Bearer <your-token-or-hive_hub_user-cookie-value>"
Claude Desktop and ChatGPT connectors
Both accept a remote MCP server over Streamable HTTP, so the same endpoint works without a local process.
Claude Desktop: Settings -> Connectors -> Add custom connector, with the URL
https://dibs.hivecommons.dev/mcp. Where the connector UI allows a custom
header, set Authorization: Bearer <token>; the read-only tools work without
one, and submit_idea needs it.
ChatGPT: Settings -> Connectors -> Add custom connector (available on plans where custom MCP connectors are enabled), using the same URL and header.
GitHub Copilot CLI and VS Code
Copilot CLI has a one-line command to add a remote MCP server:
copilot mcp add --transport http \
--header "Authorization: Bearer <your-token-or-hive_hub_user-cookie-value>" \
dibs https://dibs.hivecommons.dev/mcp
This writes to ~/.copilot/mcp-config.json using the same mcpServers / type: "http" shape as the Generic MCP JSON block below.
VS Code reads ~/.copilot/mcp-config.json (shared with Copilot CLI) and workspace .mcp.json files. Both use the same mcpServers format. For more details, see the Copilot CLI documentation and VS Code MCP server documentation.
Cursor
Cursor reads MCP configuration from ~/.cursor/mcp.json (global) or .cursor/mcp.json (project-level). Use the mcpServers format with url and headers:
{
"mcpServers": {
"dibs": {
"url": "https://dibs.hivecommons.dev/mcp",
"headers": {
"Authorization": "Bearer ${env:DIBS_TOKEN}"
}
}
}
}
Cursor supports ${env:NAME} interpolation in headers, so you can set DIBS_TOKEN as an environment variable rather than storing it in the file. For details, see the Cursor MCP documentation.
Gemini CLI
gemini mcp add --scope user --transport http \
--header "Authorization: Bearer <your-token-or-hive_hub_user-cookie-value>" \
dibs https://dibs.hivecommons.dev/mcp
This adds the server to ~/.gemini/settings.json; drop --scope user to add
it to the project's .gemini/settings.json instead. For details, see the
Gemini CLI MCP documentation.
Codex
Codex (CLI and IDE extension) reads MCP servers from ~/.codex/config.toml.
bearer_token_env_var keeps the token out of the file and sends it as
Authorization: Bearer <token>:
[mcp_servers.dibs]
url = "https://dibs.hivecommons.dev/mcp"
bearer_token_env_var = "DIBS_TOKEN"
Then export DIBS_TOKEN=<your-token-or-hive_hub_user-cookie-value> before
starting Codex. For details, see the Codex MCP documentation.
Generic MCP JSON
{
"mcpServers": {
"dibs": {
"type": "http",
"url": "https://dibs.hivecommons.dev/mcp",
"headers": {
"Authorization": "Bearer <your-token-or-hive_hub_user-cookie-value>"
}
}
}
}
Recall ideas you never acted on
Dibs ships an MCP prompt, recall_ideas, for the ideas that never made it to
a form: the ones mentioned in passing and forgotten. Dibs cannot find those
itself - only the assistant holding your history can - so the prompt asks it
to.
It instructs the assistant to search its memory and your past conversations for product ideas, feature wishes, requirements and specs you described but never acted on, report each with enough context to act on, and then ask which ones to list. It submits only what you pick.
| Argument | Purpose |
|---|---|
scope |
Optional focus - a project, a repo, a period like "the last year". Omit to search everything available. |
limit |
Maximum candidates to report. Defaults to 10, capped at 50. |
In Claude Code and Claude Desktop the prompt appears as a slash command once
the connector is added. Clients that do not surface prompts in their UI can
still call prompts/get directly.
Capture voice notes from a Plaud recorder
dibs ingest plaud turns Plaud Note / Note Pro recordings into private draft
ideas, unattended and at no cost beyond the Plaud subscription. It talks to the
official, free Plaud MCP server (npx -y @plaud-ai/mcp@latest, over stdio),
lists recordings since its cursor, fetches each one's Plaud summary and
transcript, and submits them with the submit_idea tool (tags plaud,voice-note).
The flow is: record on the Plaud → the Plaud app uploads and transcribes it →
dibs ingest plaud (one pass, or a poller) → a private draft idea appears in
Dibs, tagged plaud, with the Plaud summary and transcript as its body.
# Once: sign in to Plaud through its MCP server (the token stays in ~/.plaud).
# Then, with your hub token (see "Token" above):
export DIBS_TOKEN=...
dibs ingest plaud --dry-run # first time: see what would be submitted
dibs ingest plaud # one pass, for cron/launchd
dibs ingest plaud --interval 15m # or keep polling
First run: Plaud sign-in
The Plaud MCP server owns the Plaud login. On its first start it prints a
sign-in URL on stderr (which dibs ingest plaud forwards) and waits for the
browser to redirect to its callback on http://localhost:8199. Open the URL,
approve, and the server writes its OAuth token to ~/.plaud/tokens-mcp.json;
later runs are unattended. Dibs never reads, copies, or logs that file.
On a headless or remote host the callback still has to reach the machine running the server. Either forward the port while you sign in:
ssh -L 8199:localhost:8199 user@host # then run dibs ingest plaud there and open the URL locally
or sign in once on your laptop (npx -y @plaud-ai/mcp@latest from any MCP
client) and copy ~/.plaud/tokens-mcp.json to the host, keeping it readable
only by that user. --home (or $DIBS_PLAUD_HOME) points the server at a
different $HOME, so a service account can keep its token store under, say,
/var/lib/dibs-plaud/.plaud/ without affecting the caller's own ~/.plaud.
| Flag | Default | Purpose |
|---|---|---|
--server |
$DIBS_URL or https://dibs.hivecommons.dev |
Dibs base URL (including any base path) |
--token |
$DIBS_TOKEN |
Hub bearer token; prefer the env var, flags show up in ps. Never logged |
--interval |
0 |
Poll interval; 0 runs once |
--cursor |
~/.dibs/plaud-cursor.json |
Submitted Plaud ids, newest recording time, recent submission times |
--mcp-command |
$DIBS_PLAUD_MCP_COMMAND or npx -y @plaud-ai/mcp@latest |
Plaud MCP server command (whitespace-separated, no shell quoting) |
--since-days |
7 |
How far back the first run looks |
--tags |
plaud,voice-note |
Comma-separated tags |
--max-per-hour / --max-per-day |
10 / 30 |
Submission budget per rolling hour / 24h (0 = unlimited) |
--max |
0 |
Recordings handled per run (0 = no cap); the rest wait for the next run |
--dry-run |
off | Log each idea that would be submitted; submit nothing and leave the cursor untouched |
--home |
$DIBS_PLAUD_HOME or inherit |
$HOME for the Plaud MCP server, i.e. the parent of its .plaud/ token store |
Each recording is submitted once: its Plaud id is recorded in the cursor file
right after the submission succeeds. Recordings with no summary or transcript
yet are retried on the next run. Long transcripts are fetched page by page
(following the server's next_cursor). When the budget is used up the run
logs budget_exhausted=true; when Dibs answers HTTP 429 it logs
rate_limited=true; either way the run stops and the remaining recordings
wait for the next run. Descriptions are cut to the 64 KiB idea body limit. The helper never reads or
copies Plaud's OAuth token (~/.plaud/tokens-mcp.json).
The Plaud MCP server is third-party code that npx fetches from the registry
on every start, so it is started with an allowlisted environment rather than
the caller's: PATH, HOME (or --home), USER/LOGNAME, SHELL,
TMPDIR, TZ, TERM, locale (LANG, LC_*), XDG_*, NODE_* and
npm_config_* settings, proxy and CA variables (HTTP(S)_PROXY, NO_PROXY,
SSL_CERT_FILE/SSL_CERT_DIR), PLAUD_*, DIBS_PLAUD_*, and the Windows
basics. Anything else — DIBS_TOKEN, GITHUB_TOKEN, cloud or model-provider
keys — never reaches it, and a name containing TOKEN, SECRET, PASSWORD,
CREDENTIAL or AUTH is dropped even inside those prefixes (so
NODE_AUTH_TOKEN is not forwarded). To stop @latest from pulling an
unreviewed upstream release on each poll, pin it:
--mcp-command "npx -y @plaud-ai/mcp@<version>" (or $DIBS_PLAUD_MCP_COMMAND).
Future recording timestamps are clamped to the submission time in the cursor,
so they do not move the next listing window into the future. Existing cursors
already set in the future need their newest field corrected manually; retain
the seen entries to avoid resubmitting recordings.
Privacy
Recordings leave the device only through Plaud's own app and cloud; this
helper reads them back through Plaud's API with your Plaud login and sends
the summary and transcript text to Dibs as a private draft owned by your
hub identity. Nothing is published until you promote the draft. Audio is
never fetched. If you would rather not run an unattended poller, export the
transcript or summary from the Plaud app (TXT/Markdown) and upload it
yourself with POST /api/intake (multipart field file, same hub token,
25 MB cap, 10/h and 30/24h per identity); the same idea-shaping pipeline
turns it into a draft.
Cron (every 15 minutes):
*/15 * * * * DIBS_TOKEN=... /usr/local/bin/dibs ingest plaud >>"$HOME/.dibs/plaud.log" 2>&1
launchd (~/Library/LaunchAgents/dev.hivecommons.dibs-plaud.plist, load with
launchctl load on that file):
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0"><dict>
<key>Label</key><string>dev.hivecommons.dibs-plaud</string>
<key>ProgramArguments</key><array>
<string>/usr/local/bin/dibs</string><string>ingest</string><string>plaud</string>
</array>
<key>EnvironmentVariables</key><dict>
<key>DIBS_TOKEN</key><string>...</string>
<key>PATH</key><string>/usr/local/bin:/opt/homebrew/bin:/usr/bin:/bin</string>
</dict>
<key>StartInterval</key><integer>900</integer>
</dict></plist>
Dibs MCP questions
How do I add Dibs MCP to Claude Code?
Run claude mcp add --transport http dibs https://dibs.hivecommons.dev/mcp in your terminal. Add --scope user to make it available in every project.
How do I add Dibs MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
How can I submit ideas using an AI agent?
You can use the streamable HTTP MCP endpoint to allow clients like Claude Code, Cursor, or Copilot CLI to submit ideas directly.