agentsclimarketplace

Setup

Skill franzos/claude-plugins/plugins/stackpit/skills/setup

Claude Code plugin marketplace: a set of domain experts (subagents) grouped into installable plugins.

Install
npx -y skills add franzos/claude-plugins --skill setup

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 28 days oldThe repository was created 28 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use to set up a new Stackpit deployment (the self-hosted, single-binary drop-in for Sentry event ingestion), with or without Docker and on SQLite or Postgres. Runs the init template, sets the load-bearing config (bind addresses, external URLs, admin token, master key, filter mode), points SDKs at the ingest listener, and verifies the instance boots and accepts events. Triggers: installing Stackpit, first-time setup, standing up Stackpit, getting Stackpit running, self-hosting Sentry replacement.

SKILL.md

5.0 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it

Setting up Stackpit

Stackpit is a single Rust binary backed by one SQLite file (Postgres optional), with two HTTP listeners in one process: an admin listener (web UI + JSON API, default 127.0.0.1:3000) and an ingestion listener (SDK traffic, default 0.0.0.0:3001). Do not hardcode the schema from memory: docs/operator-guide.md and the stackpit init template are the source of truth. Read them from the deployment you are working on.

Work through these in order. Confirm the choice at each fork with the user rather than guessing.

1. Locate the deployment and pick a shape

Find what you are working with: a repo checkout (docs/operator-guide.md, README.md, SCALING.md), a package install (cargo install stackpit, Homebrew, .deb/.rpm, Guix), or a container image (ghcr.io/franzos/stackpit, or -postgres). Then pick the shape and confirm it:

  • Docker: single multi-stage Dockerfile, backend chosen at build via ARG DB_FEATURE=sqlite|postgres; two published image variants (:latest SQLite, :latest-postgres). EXPOSE 3000 3001, healthcheck on :3001/health. There is no docker-compose.yml in the repo, only Dockerfile guidance; generating one is net-new, do it only if the user wants it.
  • Native: install via package or cargo build --release. Backends are mutually-exclusive cargo features (--no-default-features --features sqlite or --features postgres; default is sqlite). Guix dev toolchain is manifest.scm + .envrc.

Read the install and first-boot sections of README.md and the config reference in docs/operator-guide.md before proceeding.

2. Provision the toolchain (native builds only)

If building from source, get a toolchain without touching the host: invoke the provision-environment skill (prefer the project manifest.scm via guix shell; Rust 1.88+). Skip for package or container installs.

3. Initialise the config

Run stackpit init to write stackpit.toml (default path; override with -c/--config <path>) with a random admin token. Set at minimum:

  • [server].bind / [server].ingest_bind — admin binds loopback by default; in a container you must set bind = "0.0.0.0:3000" for the mapped admin port to be reachable. Keep ingest reachable by SDKs (0.0.0.0:3001 is intended and public).
  • [server].external_url / external_ingest_url — the URLs clients and SDKs actually reach (behind any proxy).
  • [server].admin_token — min 16 chars, superuser/break-glass above all orgs. Generate a strong one (openssl rand -hex 32); do not ship the dev value.
  • [server].master_key (or the STACKPIT_MASTER_KEY env var, which overrides it) — 32-byte hex key for at-rest secret encryption. Required if you enable OAuth; without it, integration secrets (Slack tokens, webhook URLs) fall back to plaintext in the DB with a warning. Keep it out of the DB's directory and backups.
  • [storage] — path for SQLite, or database_url for Postgres (overrides path); retention_days (default 90; 0 never expires).
  • [filter].mode — open auto-provisions projects on first ingest (convenient, permissive); closed requires pre-registration. For an exposed instance, prefer closed with rate_limit and max_projects bounds.
  • [email] (optional) — provider (lettermint/postmark/sendgrid/smtp) and credentials for instance-wide mail.

For SSO, org mapping, and alerts, hand off to stackpit:reconfigure rather than wiring them here.

4. Boot and connect an SDK

Start with stackpit serve (or serve --ingest-only for an ingest-only node). Then point an existing Sentry SDK at the ingest URL using the Stackpit DSN (format in docs/operator-guide.md §Connecting SDKs).

5. Verify before declaring done

  • Confirm health: GET :3001/health.
  • Log in to the web UI on the admin listener with the admin token.
  • Send a test event from an SDK (or stackpit-bench) and confirm it appears as an issue. Do not report setup as working without seeing an event land.

6. TLS and reverse proxy

Stackpit serves plain HTTP; terminate TLS at a reverse proxy (nginx/Caddy). If auth is enabled and the bind is non-loopback, set [server].force_secure_cookies = true and an https:// external_url, or startup fails. Set [server].trusted_proxies to the proxy's IP/CIDR so X-Forwarded-For is trusted correctly (loopback is always trusted; misconfiguration enables rate-limiter IP spoofing).

For scaling the ingest path (Postgres, ingest_writers, ingest_batch_size), see SCALING.md.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most project setup skills give in ~1.2k tokens

Counted across 999 of the 1,637 authors here whose files we hold, read 2026-08-07

  • Ask one question at a timein 29 of 999, across 28 files
  • Detect the package manager from lockfilesin 28 of 999, across 9 files
  • Present findings to the userin 26 of 999, across 5 files
  • Explore current repo statein 24 of 999, across 3 files
  • Update the agent skills block in place if it existsin 24 of 999, across 3 files
  • Install husky lint-staged and prettierin 23 of 999, across 4 files
  • Create the lintstagedrc filein 22 of 999, across 3 files
  • Commit all changed filesin 22 of 999, across 3 files
  • Run lint-staged to verify it worksin 22 of 999, across 3 files
  • Create the husky pre-commit filein 21 of 999, across 2 files
  • Create a prettierrc file if missingin 21 of 999, across 2 files
  • Initialize huskyin 21 of 999, across 2 files

Said here and by no other author read

  • read the operator guide and init template
  • confirm the deployment shape with the user
  • invoke provision-environment for native builds
  • run stackpit init
  • set load-bearing config values
  • generate a strong admin token

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.