Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/frankxai/Starlight-Intelligence-SystemWrote 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/commands/frankxai/starlight-intelligence-system/culture-rituals)<a href="https://agentmods.dev/commands/frankxai/starlight-intelligence-system/culture-rituals"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/culture-rituals/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/frankxai/starlight-intelligence-system/culture-rituals"><img src="https://agentmods.dev/badge/commands/frankxai/starlight-intelligence-system/culture-rituals.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00083 | $0.03092 |
| Opus 5 | $0.00042 | $0.01546 |
| Sonnet 5 | $0.00017 | $0.00618 |
| Haiku 4.5 | $0.00008 | $0.00309 |
Grade A, and why
culture-rituals 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 12d 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 — 226 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/culture-rituals
Load SIP.md, VOICES.md, agents/starlight-culture.md, skills/people-intelligence/culture-design.md. Produce a Ritual Architecture.
Refusal frame
This command does not produce off-site agendas, awards-ceremony plans, or single-event culture initiatives. Rituals are weekly/monthly/quarterly recurrences that compound. Annual events are spikes, not architecture. If the request is "design our annual all-hands theme," route out.
Input
$ARGUMENTS
Flags
--size <S|M|L>— S = under 50, M = 50-250, L = 250+. Ritual cadence and facilitation pattern shifts by size.--pattern <onsite|hybrid|remote>— physical pattern affects which rituals work and which collapse. Required.--values "v1,v2,v3"— the operationalized values that rituals must reinforce. Required.- Optional context paragraph — current rituals (if any), current pain points, recent restructure.
Process
- Audit current rituals + anti-rituals — what does the org already do across the four cadences? Which create belonging, which erode it?
- Design rituals across four cadences + annual milestones — each ritual SCARF-grounded, belonging mechanism named, success indicator specified.
- Anti-ritual removal recommendations — surface current anti-rituals (back-to-back meetings, recognition-by-visibility, "open door" theater, all-hands as broadcast) and recommend their removal before adding new rituals.
- Facilitation protocol per ritual — who runs it, in what format, with what time-box, with what artifact.
- Save — write to
people-intelligence/culture/rituals-<org>-<date>.md. - Hand off — exactly one named next move.
Output format
# Ritual Architecture — <Org Name> — <YYYY-MM-DD>
> *"Rituals create belonging through shared, repeated experience (Turner, Bell). Most companies have weak rituals (annual all-hands, awards) and strong anti-rituals (back-to-back meetings, async-everything that erodes weak ties). The architecture is what's missing."*
## Context
- **Org:** <name>, <size>, <pattern>
- **Operationalized values to reinforce:** <list>
- **Current rituals (audited):** <list>
- **Current anti-rituals (audited):** <list>
---
## Section 1 — Anti-ritual audit (remove before adding)
The most overlooked move. Surface current anti-rituals and recommend removal **before** layering new rituals on top.
| Anti-ritual | Mechanism of harm | SCARF threat activated | Belonging impact | Remediation |
|-------------|-------------------|------------------------|------------------|-------------|
| Back-to-back meeting culture | No recovery → cortisol-driven decisions, no deep work | Autonomy, Certainty | Erodes individual capacity → weakens collective output | 25/50-minute defaults; meeting-free days; explicit no-meeting-after-hours; calendar audit at manager level |
| Recognition-by-visibility | Rewards extroverts and self-promoters; punishes deep workers | Status, Fairness | Builds resentment; signals that "being seen" matters more than contribution | Replace with peer-nomination + manager-review hybrid; surface non-visible contributions explicitly |
| All-hands as broadcast | One-way info dump; zero genuine Q&A → activates Status threat | Status, Autonomy | Reinforces passive culture | Replace 30 min broadcast with 10 min broadcast + 20 min real Q&A or breakout dialogue |
| "Open door" policy theater | Punishes those who use it (perceived as not autonomous) | Autonomy, Status | Silent disagreement | Replace with structured 1:1 cadence + explicit "bad news welcome" framing in 1:1 charters |
| Async-everything for distributed teams | Erodes weak-tie relationships → social distance activates exclusion circuits (Eisenberger) | Relatedness | Belonging collapses; engagement drops | Add scheduled real-time rituals (weekly team rhythm, monthly virtual coffee) — async + sync hybrid |
| <other anti-ritual> | <mechanism> | <SCARF> | <belonging> | <remediation> |
---
## Section 2 — Daily rituals
Short, low-cost, high-frequency. Build the substrate.
### Daily Ritual A: <name — e.g., "Async standup post">
- **Purpose:** team coordination + lightweight visibility
- **SCARF activation:** Certainty (clarity on what's happening), Relatedness (lightweight presence)
- **Belonging mechanism:** small, repeated co-presence
- **Format:** 3-line post (yesterday / today / blockers) in shared channel by <time>
- **Time-box:** ≤ 5 minutes per person
- **Facilitator:** rotating team-member; no manager-led
- **Success indicator:** ≥ 80% participation rate; signal-to-noise stays high; blockers actually surface
- **Avoid:** turning it into status-reporting theater; manager-led format (kills the belonging mechanism)
### Daily Ritual B: <if applicable for the size/pattern>
<same structure>
---
## Section 3 — Weekly rituals (the substrate of belonging)
Most-load-bearing cadence. The weekly rhythm is where the lived culture is actually expressed.
### Weekly Ritual A: <name — e.g., "Friday wins + lessons">
- **Purpose:** belonging through shared celebration + psychological safety through normalized lesson-sharing
- **SCARF activation:** Status (recognition), Relatedness (shared experience), Fairness (everyone gets a turn)
- **Belonging mechanism:** Turner-style ritual through repeated shared experience
- **Reinforces value:** <which operationalized value(s)>
- **Format:** 30-minute team meeting; round-robin (each person shares one win + one lesson); manager goes last
- **Facilitator:** rotating; manager goes last specifically (Edmondson — manager-first kills psychological safety in subsequent rounds)
- **Success indicator:** ≥ 90% attendance; lessons shared include genuine failures (not curated wins-disguised-as-lessons)
- **Avoid:** manager-led; performative wins-only; skipping when busy (signals it doesn't matter)
### Weekly Ritual B: <name — e.g., "Manager 1:1">
- **Purpose:** psychological safety + individual context + early surfacing of friction
- **SCARF activation:** Relatedness (one-on-one attention), Autonomy (private space to disagree)
- **Reinforces value:** <which>
- **Format:** 30-45 min, weekly, walking or seated, agenda owned by report (not manager)
- **Time-box:** 30-45 min
- **Facilitator:** report drives agenda; manager listens 70% of the time
- **Success indicator:** report regularly brings difficult topics; manager hears bad news early not late
- **Avoid:** status-update meetings disguised as 1:1; manager-driven agenda; cancellation pattern
### Weekly Ritual C: <if applicable>
<same structure>
---
## Section 4 — Monthly rituals (milestone markers)
Lower frequency, higher meaning. Mark progress + reinforce values.
### Monthly Ritual A: <name — e.g., "Peer-recognition ceremony">
- **Purpose:** value-reinforcement via public recognition
- **SCARF activation:** Status (positive recognition), Fairness (peer-nominated reduces favoritism), Relatedness (shared celebration)
- **Reinforces value:** <which operationalized value(s)>
- **Format:** 20 min, anyone can nominate anyone in advance; named recognition with specific behavior + which value it exemplifies
- **Time-box:** 20 min
- **Facilitator:** rotating; CEO does not run this
- **Success indicator:** broad nomination distribution (not the same 5 people every month); recognitions cite specific behavior + value
- **Avoid:** manager-only nominations; vague recognitions ("great job!"); same names recycled
### Monthly Ritual B: <e.g., "Learning showcase">
<same structure>
---
## Section 5 — Quarterly rituals (meaning-making)
Strategic-cadence rituals. Reinforce direction, surface drift, celebrate compound progress.
### Quarterly Ritual A: <name — e.g., "Quarterly retrospective">
- **Purpose:** meaning-making + system-level retrospective + drift detection
- **SCARF activation:** Certainty (clarity on direction), Autonomy (team agency over the next quarter), Fairness (transparent reflection)
- **Reinforces value:** <which>
- **Format:** half-day off-cycle; team-by-team retro followed by cross-team synthesis; CEO present in synthesis only
- **Time-box:** 4 hours
- **Facilitator:** ops or chief-of-staff; outside facilitator for orgs >100
- **Success indicator:** specific system-level changes committed; manager + IC voice present; retro outputs actually drive Q+1 OKRs
- **Avoid:** retro-as-vibes; CEO-led synthesis (kills psychological safety); retros that produce no committed change
### Quarterly Ritual B: <e.g., "Strategy share">
<same structure>
---
## Section 6 — Annual milestones (spikes, not architecture)
Annual events mark time but do not constitute the cultural substrate. Use sparingly and with purpose.
| Milestone | Purpose | Format | What it does NOT do |
|-----------|---------|--------|---------------------|
| Founding day | Origin-story reinforcement | Founder narrative + early-team reflection | Replace weekly rituals; pretend culture is built once a year |
| Annual all-hands | Strategic clarity + vision reinforcement | 60-90 min broadcast + 60 min real dialogue | Substitute for weekly team rhythm |
| Year-in-review | Compound-progress recognition | Data + named contribution stories | Surface yearly performance gaps that should have been addressed quarterly |
| Anniversary recognition | Tenure as belonging signal | Named, specific, contribution-cited | Become a participation trophy |
---
## Section 7 — Distribution map
Which rituals are critical for which roles / locations / patterns?
| Cadence | Onsite | Hybrid | Remote-first |
|---------|--------|--------|--------------|
| Daily | Async standup post | Async standup post + 2x/week in-person sync | Async standup + scheduled video presence 1x/week |
| Weekly | Friday wins + 1:1 | Friday wins + 1:1 + structured async-to-sync handoff | Friday wins (video) + 1:1 + virtual coffee rotation |
| Monthly | Peer-recognition + learning showcase | Same + intentional in-person convening | Same + quarterly in-person gathering substitute |
| Quarterly | Retrospective + strategy share | Same | Same — anchored in scheduled video, supported by written-first context |
---
## Section 8 — Implementation sequence (8 weeks)
- **Weeks 1-2** — Anti-ritual removal (do not add new rituals on top of unkilled anti-rituals)
- **Weeks 3-4** — Weekly rituals launched (substrate first)
- **Weeks 5-6** — Daily + monthly rituals layered
- **Weeks 7-8** — Quarterly + annual baseline established
- **Quarter 2+** — Re-measure psychological safety, attendance, broad nomination distribution
---
## Load-bearing next move
**`/culture-onboarding-90 <role>`** — onboarding is the entry point for every ritual; new hires must encounter the architecture in their first 90 days or they will not internalize it. Default next.
Alternatives:
- `/culture-values-ops <org>` — if rituals revealed value-system gaps (rituals can't reinforce values that aren't operationalized)
- `/culture-design <org>` — if anti-ritual audit revealed audit-level gaps not previously surfaced
---
**Built on SIP** — Starlight Intelligence Protocol
- Substrate: starlightintelligence.org/protocol v1.1.0
- Layers used: [file-contract, attestation, commands, sovereignty]
- Verticals: [email protected] (People Intelligence — Culture, sub-system 4 of 6)
- Generated: <ISO date>
---
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.
- 12d ago First seen · 226 lines · 83 tokens per session scan A a66bddec62c9
culture-rituals is a command published in the GitHub repository frankxai/Starlight-Intelligence-System (8 stars, last pushed today), licensed MIT. It adds 83 tokens to every session and 3,092 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 commands, from other repositories
memories
View and manage learned memories.
mpm-session-resume
Load context from paused session.
forget
Delete specific memories.
learn
Add new learning to memory.
gbu-retro
Post-session retrospective — harvest this session's lessons into durable doctrine.
consolidate
Write a compact checkpoint summary of the current frontier.