From loom
Use Loom's shared record grammar. Use when creating, naming, linking, validating, repairing, or reconciling Loom artifacts.
npx claudepluginhub z3z1ma/agent-loom --plugin loomThis skill uses the workspace's default tool permissions.
This skill is the shared grammar layer for Loom artifacts.
references/change-class.mdreferences/claim-coverage.mdreferences/frontmatter.mdreferences/implementation-reality.mdreferences/naming-and-ids.mdreferences/packet-frontmatter.mdreferences/query-and-linking.mdreferences/repair-and-drift.mdreferences/retrospective.mdreferences/route-vocabulary.mdreferences/semantic-link-usage.mdreferences/status-lifecycle.mdreferences/validation.mdPreserve observed artifacts as evidence. Use when validation outputs, reproductions, logs, scans, screenshots, or red/green results need claim links.
Provides template patterns, maturity lifecycles, and naming conventions for knowledge storage, documentation systems, and configuration management.
Guides Next.js Cache Components and Partial Prerendering (PPR): 'use cache' directives, cacheLife(), cacheTag(), revalidateTag() for caching, invalidation, static/dynamic optimization. Auto-activates on cacheComponents: true.
Share bugs, ideas, or general feedback.
This skill is the shared grammar layer for Loom artifacts.
It is the place to consult when you need to know:
A Loom record should be legible to both:
That means:
loom-evidence for the evidence record and
return here only for shared naming, linking, or validation grammarPrefer records that are:
templates/ directory, when one existsRead immediately for any non-trivial record creation or repair:
references/frontmatter.md when creating or repairing record metadata.references/packet-frontmatter.md when creating or repairing packet
frontmatter for Ralph, critique, or wiki packet support artifacts.references/naming-and-ids.md when choosing IDs, filenames, or reference
shapes.Then read conditionally:
references/status-lifecycle.md when setting or auditing non-ticket
statuses.references/claim-coverage.md when requirements, acceptance, evidence, or
critique need traceable claim IDs.references/query-and-linking.md when searching graph edges, orienting in
tickets and packets, tracing claims through evidence or critique, scanning
placeholders, or reconciling references.references/semantic-link-usage.md when deciding whether a relationship
belongs in links:, depends_on, coverage, evidence, critique, or
external refs.references/implementation-reality.md when code, tests, specs, and
evidence need their truth boundaries separated.references/change-class.md when tickets or packets need evidence,
critique, or verification defaults.references/validation.md when checking structural record health.references/repair-and-drift.md when graph drift needs classification or
safe repair.references/route-vocabulary.md when deciding which skill or owner layer to
use from record truth, or when distinguishing statuses, packet outcomes,
commands, and support cues from workflow choices.references/retrospective.md when assimilating durable learning into owner
layers.skills/loom-evidence/SKILL.md when creating or validating evidence artifacts.