Agent Skills by agm1968
Showing 16 vetted skills indexed across 1 GitHub repositories.
AGM1968breadcrumbs
Capture goals, state, decisions, and next steps into node metadata.
AGM1968wv-detect-loop
Detects repeated failures and interrupts retry loops in AI-driven workflows.
AGM1968pre-mortem
Identify failure scenarios and mitigation plans for significant tasks.
AGM1968weave-audit
Audit Weave graphs for integrity issues and generate remediation reports.
AGM1968weave
Orchestrate AI coding workflows across intake, context, execute, and close phases.
AGM1968wv-verify-complete
Select and run the strongest verification method, then record evidence in node metadata.
AGM1968ship-it
Define 3-5 binary done criteria and non-goals for weave workflow nodes.
AGM1968wv-guard-scope
Enforce task scope and codebase conventions during software implementation.
AGM1968wv-decompose-work
Break down an epic into features and tasks with dependencies.
AGM1968sanity-check
Identify and tag assumptions in plans with verification statuses.
AGM1968fix-issue
Resolve GitHub issues and Weave nodes through procedural gate checkpoints.
AGM1968resolve-refs
Extract cross-references from documents and code into follow-up plans.
AGM1968wv-clarify-spec
Clarify vague task requirements using 5W+C analysis into concrete specifications.
AGM1968close-session
Save, sync, and push end-of-session work in Weave-managed repositories.
AGM1968plan-agent
Audit Git commits to generate PLAN.md and enrich Weave graph nodes.
AGM1968zero-in
Enforce a bounded search protocol with defined targets and stepwise reporting.