vitala89
UserVendor-neutral local framework and CLI for defining, validating, and synchronizing engineering intent across AI coding tools. Offline-first and non-destructive.
Categories
Indexed Skills (21)
aif-extension-review
Review an external skill, plugin, MCP server, or workflow pack for provenance, license, compatibility, capabilities, and unsafe automation before adoption or update. Use before external extension content becomes trusted or active.
aif-feature-discovery
Turn an ambiguous product or engineering idea into a reviewable feature brief through a focused interview and alternatives analysis. Use before specification or planning when new behavior, constraints, or success criteria are unsettled.
aif-task-router
Classify a non-trivial engineering request and recommend the smallest safe Intentloom workflow and skill set before planning or implementation. Use when the task route, required discovery, or review depth is not already explicit.
aif-verification-gate
Verify a completed change against its acceptance criteria and repository checks before any completion, commit, or pull-request claim. Use at the end of implementation, bug fixing, refactoring, or generated-output updates.
aif-branch-finisher
Prepare a completed branch for human handoff without external publication.
aif-commit-writer
Draft a concise commit message for an already reviewed change.
aif-context-pack
Assemble minimal cited context for a bounded engineering task.
aif-docs-sync
Synchronize documentation with a verified behavior or interface change.
aif-iteration-handoff
Write a compact handoff for the next bounded iteration.
aif-orchestrator
Coordinate a bounded Intentloom engineering iteration when the user asks to plan or sequence work.
aif-privacy-review
Review data handling for collection, retention, exposure, and minimization concerns.
aif-project-adoption
Assess safe Intentloom adoption for an existing repository.
aif-project-state-sync
Synchronize canonical project state documentation after verified changes.
aif-security-review
Review a bounded change for the Intentloom threat model risks.
aif-specification-review
Review a feature brief for testable behavior and missing decisions.
aif-technical-debt-sync
Record or retire explicit technical debt after a verified decision.
aif-code-review
Review a bounded diff independently against repository standards and the originating specification, then report correctness and verification gaps. Use for branch, pull-request, or work-in-progress review without modifying code.
aif-debugger
Diagnose an observed defect or performance regression through a reproducible feedback loop, ranked hypotheses, and regression evidence. Use when behavior is broken, failing, inconsistent, or unexpectedly slow.
aif-planning-review
Review an implementation plan for scope, dependencies, architecture, risk, rollback, and verification gaps. Use after feature discovery or specification and before implementation begins.
aif-testing-strategy
Design proportionate tests and feedback loops for specified behavior, including red-green-refactor where it provides reliable evidence. Use when acceptance criteria and existing test conventions are known.
aif-feature-builder
Implement one approved, bounded feature slice.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.