agentleFS
bks-labbks-lab/open-bridge/skills/bridge-greeting/SKILL.md

bridge-greeting

Skill13 starsChanged 13 days agoMITView on GitHub

Context-aware terminal greeting (MOTD) for Bridge instances: renders a per-instance logo, palette, today's calendar and the "Doing" board, resolved from which org folder the terminal opened in. Includes gen-logo.py (ANSI-Shadow generator). Trigger: "/bridge-greeting", "terminal greeting", "motd", "startup logo", "instance logo", "fastfetch logo", "folder logo".

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    When the resolved instance's bridge-config.yaml has a non-empty purpose.statement, render it as the subtitle / tagline directly under the …

    line 40
  2. Very specific:

    Write the chosen values into that instance's theme branding: block (or the machine-local override).

    line 77

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Bridge Greeting

A login-shell greeting that is instance-aware: the org folder you open the terminal in resolves to a Bridge instance, and the greeting then shows that instance's identity — logo, colours, calendar tags, task board, and repos.

Home / unknown dirs → the full system dashboard (fastfetch cascade + global board). Inside an org tree → a focused panel for the resolved instance.

The model — branding belongs to the Bridge, not the script

The greeting reads its visual identity from the resolved instance's theme (themes/<theme>.yaml → branding: block), so a logo "comes from" its Bridge instance the same way identity/agent/ and vocabulary do. Three layers:

LayerOwnsTier
Engine (scripts/render-motd.sh)discovery, rendering, board parsing — no instance literalsCORE → open-bridge
Theme branding (themes/<theme>.yaml branding:)logo path, logo_color_1/2, calendar_tags per instanceper-instance
Local override (~/.config/fastfetch/bridge-motd.local.sh)per-org fallbacks for instances whose theme has no branding: yet; the org→instance machine mapUSER (machine-local, never committed)

Discovery scans ~/Developer/*/*/bridge-config.yaml to build the org-folder → instance map automatically — no hardcoded instance names in the engine.

Purpose tagline — the instance's north-star under the logo

When the resolved instance's bridge-config.yaml has a non-empty purpose.statement, render it as the subtitle / tagline directly under the logo, alongside the theme-derived identity the greeting already shows — so the terminal opens reading as what this Bridge is for, not a bare wordmark. Empty purpose.statement → no tagline line (today's behaviour). This is display only; the purpose never changes what the greeting can show.

gen-logo.py — make an instance its logo

uv run --with pyfiglet python scripts/gen-logo.py --top ACME --bottom CORP > assets/logos/acme.txt
uv run --with pyfiglet python scripts/gen-logo.py --top ACME --umlaut 1   > assets/logos/acme.txt

--umlaut N re-adds a diaeresis over the N-th letter of --top (figlet drops non-ASCII glyphs, so an umlaut would otherwise be lost). Output carries fastfetch $1/$2 colour placeholders; the actual colours are applied at render time from the theme's logo_color_1/2.

--pipe false is mandatory when fastfetch output is captured — a plain --pipe strips every SGR code, leaving the logo uncoloured.

Setup — advise, don't assume

Branding is taste. When setting up (or reconfiguring) an instance's greeting — during onboarding or on /bridge-greeting — consult the user, propose, let them decide (Bridge house style: advise, don't act). Ask, per instance:

  • Wordmark + subtitle — top word, optional second line (e.g. a unit/role).
  • Colour mode — one of:
    • monochrome — no colour, renders in the terminal's own foreground (set logo_color_1/2: none). For users who want a plain, uncoloured logo.
    • single — one uniform colour, no two-tone alternation (logo_color_1 == logo_color_2).
    • two-tone — distinct logo_color_1/logo_color_2 (gradient feel).
  • Calendar tags — which calendars surface for this instance.

Write the chosen values into that instance's theme branding: block (or the machine-local override). Never hardcode a palette — default to asking.

Deploy (thin-shim pattern)

scripts/render-motd.sh is the single source of truth. .zshrc calls a one-line shim at ~/.config/fastfetch/bridge-motd.sh that execs the repo engine — so it auto-follows git pull, no copy to re-deploy:

ENGINE="$HOME/Developer/<org>/<your-bridge>/skills/bridge-greeting/scripts/render-motd.sh"
[[ -r "$ENGINE" ]] && exec bash "$ENGINE" "$@"   # silent skip if repo absent → login never fails

Per-machine data stays out of the repo: logo files in ~/.config/fastfetch/logos/, per-org overrides in ~/.config/fastfetch/bridge-motd.local.sh (USER, never committed).

Two renderers, one concept (macOS + Windows)

The greeting is platform- and provider-pluggable. The same three-layer model (generic engine · instance config · branding asset) has two engines:

macOS armWindows arm
Enginescripts/render-motd.sh (bash + fastfetch)scripts/render-greeting.ps1 (PowerShell + ANSI)
Data sourceGitHub Projects V2 (gh)Azure DevOps — WIQL / pipelines / commits (az)
Instance configbridge-motd.local.sh (BRIDGE_MOTD_*)$BridgeGreetingConfig hashtable
Brandingtheme branding: blocksame assets/logos/*.txt + colour
Deploy.zshrc shim → repo engine$PROFILE shim → dot-source engine + config

render-greeting.ps1 is generic — no instance literals. A profile dot-sources it plus an instance config, then calls Show-BridgeGreeting -Config $cfg. The config is a hashtable: Label, Logo (a assets/logos/*.txt reused as-is — the PS renderer decodes the same $1/$2 + \u{} format), Color1/Color2, Provider, an Ado block (Org, Project, RepoId, ResourceId, Branch), and an ordered Sections array (Kind ∈ workitems/pipelines/commits, with per-section Wiql, Title, Empty). No config → it greets with just the header. Results cache to JSON (default 1 h TTL).

Provider = 'azure-devops' is implemented; github is a documented extension point (add a Get-GhSection dispatch — the rest of the engine is provider-agnostic).

Windows deploy mirrors the thin-shim idea but dot-sources by $PSScriptRoot (a dev-box typically has no Bridge clone, so the engine + config + profile ship as siblings in the PowerShell profile folder, e.g. a cloud-synced one):

# Microsoft.PowerShell_profile.ps1 (thin shim)
. "$PSScriptRoot\render-greeting.ps1"
. "$PSScriptRoot\bridge-greeting.config.ps1"   # defines $BridgeGreetingConfig
Show-BridgeGreeting -Config $BridgeGreetingConfig

The instance config + its logo + the deployed profile are USER scope and live in that instance's own Bridge repo, never here — exactly like the macOS local override. The engine is the only CORE/promotable piece.

Branding everywhere

branding.logo_* is the terminal rendering of one brand identity that also feeds the web UIs (bridge-deck, the :8790 ops dashboard, the :8793 control center, bridge-explorer). The shared SoT for web surfaces is DESIGN.md (palette + wordmark tokens). See references/architecture.md.

Layering (core / org overlay / instance)

  • open-bridge: engine + gen-logo.py + a generic OPEN BRIDGE logo + neutral palette. Discovery, no literals.
  • your org overlay (<your-org>/<your-bridge>): a sample logo + palette as the default branding:.
  • per-instance (USER): each instance's own logo, org→instance map, calendar tags — local, never promoted.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.