Agent-almanac brahma-bhaga

install
source · Clone the upstream repo
git clone https://github.com/pjt222/agent-almanac
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/pjt222/agent-almanac "$T" && mkdir -p ~/.claude/skills && cp -r "$T/i18n/caveman-lite/skills/brahma-bhaga" ~/.claude/skills/pjt222-agent-almanac-brahma-bhaga && rm -rf "$T"
manifest: i18n/caveman-lite/skills/brahma-bhaga/SKILL.md
source content

Brahma Bhaga

Generative creation from void or ambiguity — structured emergence of new patterns, approaches, and solutions where none existed before.

When to Use

  • After
    shiva-bhaga
    dissolution has cleared stale patterns and created space
  • Facing a genuinely novel problem with no obvious template or precedent
  • The user's request requires invention rather than retrieval or adaptation
  • Multiple possible approaches exist and none has been chosen — the creative act is the choice itself
  • A blank slate: new file, new project, new architecture, new approach
  • When incremental modification has reached its limits and a fresh design is needed

Inputs

  • Required: The creative goal or void to fill (available from conversation context)
  • Optional: Constraints that bound the creation (user requirements, technical limitations)
  • Optional: Seeds — fragments, inspirations, or partial ideas that inform the creation
  • Optional: What was dissolved (
    shiva-bhaga
    output) — understanding what failed guides what to create

Procedure

Step 1: Survey the Void

Before creating, understand the space available for creation.

Creative Space Assessment:
+---------------------+---------------------------+------------------------+
| Dimension           | Questions                 | Determines             |
+---------------------+---------------------------+------------------------+
| Constraints         | What MUST the creation    | The boundary within    |
|                     | satisfy? What is non-     | which creativity       |
|                     | negotiable?               | operates               |
+---------------------+---------------------------+------------------------+
| Freedom             | What is NOT specified?     | The degrees of freedom |
|                     | Where does the user leave | available for creative |
|                     | room for creative choice? | choice                 |
+---------------------+---------------------------+------------------------+
| Seeds               | What fragments, partial   | The starting material  |
|                     | ideas, or inspirations    | that informs but does  |
|                     | already exist?            | not dictate            |
+---------------------+---------------------------+------------------------+
| Anti-patterns       | What was tried before and | The space to avoid —   |
|                     | failed? What approaches   | creation that repeats  |
|                     | were dissolved?           | dissolved patterns     |
+---------------------+---------------------------+------------------------+
| Context             | What exists around the    | The environment the    |
|                     | void? What must the       | creation must fit      |
|                     | creation integrate with?  | into                   |
+---------------------+---------------------------+------------------------+
  1. Map each dimension honestly — especially constraints, which are often implicit
  2. Note the degrees of freedom: these are where genuine creation happens
  3. Identify seeds without committing to them — they inform, not dictate

Got: A clear picture of the creative space: bounded by constraints, informed by seeds, and opened by degrees of freedom.

If fail: If the space feels fully constrained (no degrees of freedom), re-examine — often constraints that seem fixed are preferences. Ask the user if needed.

Step 2: Generate — Divergent Exploration

Produce multiple possibilities without evaluating them.

  1. Generate at least three distinct approaches to filling the creative space
  2. Each approach should be genuinely different — not variations on a theme
  3. For each approach, capture:
    • The core idea in one sentence
    • How it satisfies the constraints
    • What makes it distinct from the others
    • What it sacrifices or trades off
  4. Include at least one approach that feels unconventional or risky
  5. Do not evaluate yet — generation and evaluation are separate phases

Got: Three or more genuinely distinct approaches, each with a clear identity and trade-off profile.

If fail: If all approaches feel similar, the generation was too narrow. Return to Step 1 and look for unexplored degrees of freedom. Alternatively, invert a constraint: "What if I did the opposite of the obvious approach?"

Step 3: Evaluate — Convergent Selection

Assess the generated approaches against the creative space.

  1. For each approach, assess:
    • Constraint satisfaction: Does it meet all non-negotiable requirements?
    • Elegance: Is it the simplest solution that works?
    • Resilience: Will it survive future perturbation?
    • Integration: Does it fit naturally with the surrounding context?
    • Novelty: Does it bring something genuinely new, or merely rearrange the old?
  2. Eliminate approaches that violate hard constraints
  3. Among remaining approaches, choose based on the user's implicit values (simplicity? thoroughness? creativity?)
  4. If two approaches are equally strong, present both to the user with trade-offs clearly stated

Got: A single chosen approach (or a clearly framed choice for the user) with articulated reasoning.

If fail: If no approach satisfies all constraints, the constraints may be contradictory. Surface the contradiction to the user rather than forcing a creation that compromises on fundamentals.

Step 4: Manifest — Bring into Form

Execute the chosen approach, giving it concrete form.

  1. Begin with the skeleton: the minimal structure that embodies the core idea
  2. Build outward from the core, adding detail as needed
  3. At each step, check: "Is this addition serving the core idea or diluting it?"
  4. Resist the urge to over-elaborate — creation is complete when nothing more can be removed
  5. Name what was created: a clear, descriptive identifier that captures its essence

Got: A concrete creation that embodies the chosen approach — code, plan, structure, or design that exists where void existed before.

If fail: If the manifestation diverges from the chosen approach, pause and re-read Step 3's selection. Drift during manifestation often indicates the selection was not fully committed to. Either recommit or re-select.

Step 5: Nurture — Protect the Nascent Creation

New creations are fragile. Protect them through their early stages.

  1. Test the creation against its constraints — does it work as intended?
  2. Identify the weakest point — where is it most likely to break?
  3. Strengthen the weakest point without over-engineering
  4. Hand off to
    vishnu-bhaga
    for ongoing preservation if the creation will persist
  5. Document the creative choices made: what was chosen, what was rejected, and why

Got: A creation that is tested, documented, and ready for sustained use.

If fail: If the creation fails its first test, assess whether the failure is in the creation or the test. If the creation is fundamentally flawed, return to Step 2 with the failure as a new anti-pattern seed.

Validation

  • The creative space was surveyed before generating ideas
  • At least three genuinely distinct approaches were generated
  • Selection was based on explicit criteria, not default instinct
  • The creation was manifested starting from its core, building outward
  • The creation was tested against its constraints
  • Creative choices were documented for future reference

Pitfalls

  • Creating before clearing: Attempting creation without prior dissolution produces new patterns contaminated by old ones. Run
    shiva-bhaga
    first if the space is cluttered
  • Single-option generation: Generating one approach and then evaluating it is not creation — it is executing the first idea. True creation requires divergent options
  • Novelty for its own sake: Creating something unconventional when a simple standard approach would serve better. Novelty is a tool, not a goal
  • Perfectionist manifestation: Polishing endlessly rather than shipping a working creation. A complete imperfect creation outperforms an incomplete perfect one
  • Unprotected creation: Manifesting something new and immediately moving on without testing or documentation leaves the creation vulnerable

Related Skills

  • shiva-bhaga
    — destruction creates the void that Brahma fills; dissolution precedes creation
  • vishnu-bhaga
    — preservation sustains what Brahma creates; handoff from creation to maintenance
  • intrinsic
    — creative engagement benefits from autonomous motivation; creation thrives in flow
  • learn
    — when creation requires knowledge not yet held, learning precedes generation
  • adapt-architecture
    — the morphic equivalent for creating new architectural patterns from existing systems