Skip to content

arminzou/dotbrain / plugin/skills/maintain-brain

maintain-brain

Verified

Capture, update, check, and clean up private project understanding. Use when preserving useful context, revising a Brain record, checking drift or broken references, or removing obsolete and redundant Brain material.

FileKindSize
SKILL.mdskill md6.4 KB
agents/openai.yamlconfig126 B
references/adr-format.mdmarkdown3.8 KB
references/context-format.mdmarkdown1.3 KB
references/design-format.mdmarkdown5.3 KB
references/dotbrain-skills.mdmarkdown2.1 KB
references/findings.mdmarkdown2.1 KB
references/writing.mdmarkdown6.5 KB

references/dotbrain-skills.md

View on GitHub
# Writing within dotbrain

The principles in [Writing for agents](writing.md) apply across visibility surfaces.
Placement and authority differ:

| Surface | Typical documents | Constraint |
| --- | --- | --- |
| Public repository | `AGENTS.md`, `RULES.md`, README, architecture and contributor docs | Useful without private access; contains no private Brain material |
| Private project Brain | `CONTEXT.md`, ADRs, designs, findings, references, explanations | Preserves project understanding, rationale, evidence, and applicability |
| User data root | Selected skills and their references | Follows the nearest authoring policy; placement alone does not activate a skill |

Public `AGENTS.md` routes to repository requirements in `RULES.md` and task-specific
references. `repo-rules` maintains those requirements and their routing. Public rationale
must stand on its own: express necessary reasoning without private paths, identifiers,
or operational state. Keep private sources and provenance in private records.

`maintain-brain` provides [vocabulary](context-format.md), [decision](adr-format.md),
[design and lifecycle](design-format.md), and [finding/reference](findings.md) semantics.
These formats describe records, evidence, and authority. A chosen workflow supplies
capture timing, writing responsibility, review, and delivery procedures. Beads owns
durable execution state; a design or explanatory document must not become a second
task tracker.

## Shapes for agent instructions

Choose structure according to the document's actual job:

| Shape | Useful structure |
| --- | --- |
| Linear procedure | Prerequisites above ordered steps; completion criteria for each non-terminal step |
| Loop | Repeatable protocol with a clear exit condition |
| Reference | Definitions, rules, examples, and conditional retrieval pointers |
| Tool driver | Branch selector, symptom-to-command guidance, and verification |

Name a terminal action for what it does: report findings, hand off an unresolved action,
or complete a record update. Add a separate terminal section only when it clarifies the
outcome. Keep conditions beside the actions they constrain.
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 196kUpdated Claude CodeCodexVerifiedInstall