skills/bdi-agents-a-soft-model-for-organisation/SKILL.md
Model organizations as BDI systems with filtered perception, discourse-shaped desires, accommodations, and abstraction-level hierarchy. Use for socio-technical coordination and organizational modeling. NOT for mechanistic workflows, individual psychology, or non-social domains.
npx skillsauth add curiositech/windags-skills bdi-agents-a-soft-model-for-organisationInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use this skill when the challenge is to model how messy organizations actually decide, coordinate, and resist change without pretending they are either fully rational machines or unknowable social fog.
You can model individuals, a recurring process, or the organization itself as the agent. The right level depends on what you need to explain, not on which level feels most intuitive.
Intentions are commitments that reduce future reconsideration. In organizations, that explains both useful focus and costly inertia.
Different roles, expertise, and incentives produce different organizational realities. Do not assume a shared objective world state just because everyone sees the same dashboard.
Organizational goals often come out of negotiation, rhetoric, and power, not from simple aggregation of individual preferences.
Coordination frequently happens through workable accommodations while disagreement persists. Designs that demand full agreement before action often misread how organizations function.
flowchart TD
A[Need organizational model] --> B{What must be explained?}
B -->|Specific actor choices| C[Model individuals or teams]
B -->|Recurring coordination pattern| D[Model the process as an agent]
B -->|Culture, identity, or strategic drift| E[Model the organization as an agent]
C --> F{Main coordination issue}
D --> F
E --> F
F -->|Different perceptions| G[Model filtered beliefs]
F -->|Conflicting priorities| H[Model discourse-driven desire formation]
F -->|Action despite disagreement| I[Design for accommodation, not consensus]
F -->|Hierarchy friction| J[Model abstraction-level boundaries]
Cue: the model treats disagreement as a temporary obstacle that must disappear before action can happen.
Fix: model accommodations and partial alignment explicitly.
Cue: the design assumes all participants observe the same facts and only differ in preferences.
Fix: give roles their own filtered belief models.
Cue: goals are treated as pre-existing values rather than outputs of negotiation and discourse.
Fix: include the goal-formation process in the model.
Cue: a process problem is analyzed only at the individual level, or a person-level conflict is forced into organization-level abstractions.
Fix: shift levels intentionally and compare what each view reveals.
Cue: a new platform is designed as if installing it will replace human cognition rather than support it.
Fix: treat technical systems as cognitive prosthetics that must fit existing organizational reasoning.
Marketing, legal, and engineering disagree on launch timing. The wrong move is to demand consensus on all facts. The right move is to map filtered perceptions, surface which desires emerge through stakeholder negotiation, and design an accommodation that permits action under persistent disagreement.
A company wants to automate incident routing. Model the process at the meso level to understand recurring coordination patterns, but drop to the team level where filtered perceptions and abstraction-level mismatch explain handoff failures.
references/soft-systems-cognitive-gap-formal-modeling.md: load when you need the high-level bridge between soft systems and BDI.references/intention-as-commitment-bounds-deliberation.md: load when inertia, commitment, or reconsideration are central.references/discourse-to-action-emergence-of-organizational-desires.md: load when organizational goals seem to be socially produced.references/accommodations-over-consensus-coordination-without-agreement.md: load when action happens despite unresolved conflict.references/hierarchies-abstraction-levels-expertise-coordination.md: load when hierarchy and abstraction mismatch are central to the failure.tools
Building resilient distributed systems with circuit breakers, retries with full-jitter exponential backoff, retry budgets (per-request 3-attempt + per-client 10% ratio per Google SRE), deadline propagation, and the cascading-failure math (4 layers × 3 retries = 64x amplification). Grounded in Resilience4j, Microsoft Cloud Patterns, AWS Architecture Blog (Marc Brooker), and Google SRE Book.
testing
Designing HTTP cache headers that work correctly across browsers, CDNs, and shared proxies — `Cache-Control` directives per RFC 9111, `stale-while-revalidate` and `stale-if-error` per RFC 5861, the Vary header for varying responses, and surrogate keys for tag-based purging. Grounded in IETF RFCs and Cloudflare/Fastly docs.
development
Use when designing or fixing a Content Security Policy on a real site, choosing between nonce-based and hash-based CSP, adding strict-dynamic, debugging "Refused to execute inline script" errors, deploying CSP in report-only mode first, configuring report-to / report-uri, or auditing an existing policy for unsafe-inline / unsafe-eval / wildcards. Triggers: "CSP blocks legitimate inline script", strict-dynamic, nonce-{RANDOM}, sha256-{HASH}, object-src none, base-uri none, frame-ancestors, Trusted Types, X-Content-Security-Policy obsolete, report-only vs enforced. NOT for general HTTP security headers (HSTS, COOP/COEP), Trusted Types deep dive, CORS configuration, or building a WAF.
tools
Choosing and operating an HTTP API versioning strategy that doesn't break clients — Stripe's date-based pinned versions, the Deprecation/Sunset header pair (RFC 9745 + RFC 8594), URI vs header vs media-type approaches, and the version-transformer pattern. Grounded in Stripe's published architecture and IETF RFCs.