KFT
Community@thelightarchitect · WA, US
Security Engineer & A.I. Practitioner with builder's mindset.
Agent Skills by KFT
Showing 25 vetted skills indexed across 2 GitHub repositories.
sherlock
Validate hypotheses against investigation evidence with confidence scoring.
validate
Validate investigative claims against seven citation formats and contradictions.
evidence
Analyze uploaded evidence bundles for TAC investigations using jq, grep, and MCP tools.
initialize
Initialize CAPPY case directories, extract log bundles, and generate inv_context.json.
handoff
Evaluate phase gates and validate inv_context.json for CAPPY handoffs.
deploy
Deploy CAPPY binaries, plugins, and pattern databases across local, GitLab, and GitHub environments.
investigate
Conduct an eight-phase investigation for XSOAR, XSIAM, and XDR issues.
curator
Registers and verifies investigation claims with file:line citations for CAPPY phases 2-7.
triage
Search pattern databases and Cortex documentation to assess Phase 2 confidence thresholds.
synthesis
Generate a structured investigation narrative linking phases 2-4 and identify validation gaps.
deliverables
Compile root cause statements, remediation steps, and investigation narratives into JIRA drafts and customer responses.
resolution
Validate investigation hypotheses against Cortex documentation, Confluence, and JIRA.
gate
Evaluate CAPPY investigation phase metrics against quality thresholds and return pass/fail status.
hypothesis
Generate and test falsifiable hypotheses for CAPPY Phase 4 investigations.
escalation
Rank escalation options for token-limited CAPPY investigations and log attempts to inv_context.json.
recon
Validate Phase 4 hypotheses against customer deployment environments using logs and tickets.
logging
Generate structured JSONL logs with correlation IDs for CAPPY investigations.
SNIFF
Analyze Rust and multi-language codebases for code smells and architecture conformance.
FETCH
Load research scoping, source selection, and decision context for MCP tooling.
SCOUT
Triage complexity and classify domain to define build scope.
GUARD
Identify and mitigate security risks in codebases and CI/CD pipelines.
SCRUM
Coordinate multi-perspective squad reviews with three-round cross-critique and Good/Gaps/Fixes reports.
HUNT
Execute plan phases with HITL gates and manifest-driven state management.
CHASE
Identify performance bottlenecks and infrastructure hotspots in software projects using MCP tools.