Anthony Chaudhary
Community@anthony-chaudhary
Agent Skills by Anthony Chaudhary
Showing 76 vetted skills indexed across 2 GitHub repositories.
ctx-overlay
Compact the active context window and prune the memory store.
quality-report
Analyze repository structure and documentation to generate a quality assessment report.
skill-lifecycle
Manage project skill lifecycles with usage telemetry and reversible archiving.
question-loop
Generate critical, contrarian, and steelman questions about repository design.
modularize
Refactors monolithic Go source code into modular, behavior-preserving units with automated verification.
appeal-score
Analyze technical documentation prose defects and generate prioritized improvement lists.
curate-cluster
Reconcile project documentation and script indices against the current file system state.
token-defaults-score
Audit agent kernel token-saving configurations against a scorecard doctrine.
steerability-score
Quantify repository steerability with growth-invariant KPIs and guide refactoring.
stability-score
Audit git-tracked repositories against stability KPIs and generate a JSON scorecard report.
skill-score
Quantifies skill-debt via nine KPI scorecard in agentic repositories requiring fak kernel environment.
claim-repro-score
Audits documentation against project artifacts to verify reproducibility of technical claims.
plan-audit
Reconcile fragmented project plan-state surfaces into a verified completion audit.
verify
Execute affected-package tests and bind results to the resolving commit.
skill-overlap
Detects redundant SKILL.md pairs in a skill pack via cosine similarity analysis.
spine-fanout
Generate follow-on GitHub issues after verifying an end-to-end spine via fak CLI.
negframe-score
Identify negatively-framed directives in agent steer-prose and enforce positive-action lexicon.
score-2x
Halve actionable debt on a scorecard surface and re-pin baselines at saturation.
persona-score
Audit repository file trees against persona-based scorecards to calculate persona-debt.
clean-skill
Audit Claude Code skill session logs to identify inefficient file-read patterns.
conflation-score
Validate provenance-honesty of reported metrics and exit summaries in the fak kernel.
study-repo
Analyze external codebases to extract technical borrows and file GitHub issues.
issue-triage
Classify and rank open GitHub issues into actionable JSON manifests.
trajectory-audit
Analyze Claude Code session transcripts to identify token-weighted cost inefficiencies and behavioral patterns.