conditional-agent-decide

A set of rules for choosing specialised coding skills when the task context calls for them.

In plain words
What is it for?
It is for deciding when to use skills such as requirements interviews, official-document research, adversarial review, UI design, browser testing, security checks, and automation guidance.
Why use it?
It helps an agent select relevant guidance for unclear requirements, unfamiliar code, interface work, security, testing, performance, and other situations.

Cursor rule for Cursor

Install

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.

agentmods
npx agentmods add rules/mamahoos/dot-files/conditional-agent-decide
Clone the repo
git clone --depth 1 https://github.com/mamahoos/dot-files

Made for: Cursor.

Per session 390 This file is loaded in full into every session.
When invoked 390 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00390 $0.00390
Opus 5 $0.00195 $0.00195
Sonnet 5 $0.00078 $0.00078
Haiku 4.5 $0.00039 $0.00039

Measured 2d ago against content hash 8cedd2a9816c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

conditional-agent-decide 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 2d 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.

home/.cursor/rules/conditional-agent-decide.mdc · 36 lines

What it actually says

Conditional Skills (Agent Decide)

Use these skills when the task context clearly requires them.

Define / Discovery

  • interview-me: Use when requirements are unclear and you need to extract intent before planning.
  • idea-refine: Use when the problem statement is vague or direction is unclear.
  • context-engineering: Use at session start, quality drops, or task switching.

Build / Validation

  • source-driven-development: Use when framework/library correctness must be aligned with official docs.
  • doubt-driven-development: Use for high-stakes or unfamiliar code where decisions need adversarial review.
  • frontend-ui-engineering: Use for user-facing UI/UX and accessibility work.
  • api-and-interface-design: Use for API contracts, module boundaries, and interface design.
  • browser-testing-with-devtools: Use for browser runtime behavior and interactive debugging.
  • security-and-hardening: Use for untrusted input, auth, secrets, or external integrations.
  • performance-optimization: Use for SLO/CWV targets or suspected performance regressions.
  • ci-cd-and-automation: Use for pipeline, checks, hooks, and automation changes.
  • deprecation-and-migration: Use for legacy removal and user/data migration paths.
  • documentation-and-adrs: Use when architecture/API decisions need explicit rationale.
  • observability-and-instrumentation: Use when adding logs, metrics, traces, or alerts.
  • shipping-and-launch: Use for rollout, release readiness, and rollback planning.
  • code-simplification: Use after correctness is established and maintainability is the goal.

Policy

  • Agent should choose these based on explicit task signals, not by default.
  • If a conditional skill is activated, mention which one and why in a short note.
Changes

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.

  1. 2d ago First seen · 36 lines · 390 tokens per session scan A 8cedd2a9816c

Subscribe to this mod's changes

conditional-agent-decide is a cursor rule published in the GitHub repository mamahoos/dot-files (3 stars, last pushed 3d ago), licensed MIT. It adds 390 tokens to every session, about $0.0019 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.