GRIDLOCK-NYC

@gridlock-nyc source repo

22 skills
0 followers
6 hours ago last updated

▌
5x5 · gridlock-nyc
Iterative research protocol — up to 5 rounds of 5 parallel research agents (25 total). Each round's findings shape the next round's questions. Final output is a synthesized implementation plan, not code. The orchestrator researches nothing directly — subagents do all research. Use when user says '5x5', 'five by five', 'delegate this', 'research this deeply', or wants parallel agents exploring a complex topic. Do NOT use for simple single-file changes or quick questions.

0
installs
▌
Organize · gridlock-nyc
Run the auto-tidy script to sort stray files from Desktop, Downloads, and home root into organized folders. Reports what was moved and shows remaining state. Use when user says 'organize', 'tidy up', 'sort my files', 'clean desktop', 'clean downloads', or 'organize my files'. Do NOT use for deleting files or emptying trash.

0
installs
▌
UI Ideas · gridlock-nyc bundle
Generate 3 novel UI upgrade ideas for AI coding agents and implement the ones that are buildable as Claude Code skills. Pure reasoning, no web search. Use when user says 'ui ideas', 'generate ui upgrades', 'new skill ideas', 'ui-ideas', or 'what else could we build'. Do NOT use for building a specific known skill (use skill-builder for that).

0
installs
▌
Hail Mary · gridlock-nyc bundle
Autonomous creative engine — finds a real-world problem nobody has solved well and builds a working product from scratch while the user is away. Creates a new repo in ~/projects/hailmary/ with working code, tests, and documentation. Use when user says "hail mary", "go build something", "YOLO mode", "surprise me", "build something novel", "creative mode", "go wild", "make something new", "autonomous build", "hail-mary", or "I'm stepping away, create something". Runs for up to an hour. Can read but NEVER edit existing repos. Do NOT use for building a specific requested feature (just ask Claude directly). Do NOT use for UI upgrade ideas (use ui-ideas).

0
installs
▌
Update Pr · gridlock-nyc
Auto-commit dirty working tree, push feature branch, and create or update the GitHub PR with an auto-generated description. One command to go from uncommitted changes to an up-to-date PR. Use when user says 'update my PR', 'update the PR', 'push and update PR', 'refresh PR', 'sync my PR', 'create a PR', 'open a PR', 'push my changes', 'get my PR up to date', 'rebase and push', 'push to the PR', 'PR update', or 'ship it'. Do NOT use for reviewing PRs, reading PR comments, or merging PRs.

0
installs
▌
Code Edges · gridlock-nyc bundle
Audit a project for edge cases, error-prone code paths, and unhandled failure modes. Researches the codebase, identifies the highest-risk areas, enumerates specific edge cases from a comprehensive catalog, writes and runs tests to prove them, then produces a prioritized risk report. Use when user says 'edge audit', 'find edge cases', 'what could break', 'edge case review', 'audit for edge cases', 'stress test this code', 'what am I missing', 'failure modes', 'where will this break', 'robustness check', or 'error path audit'. Do NOT use for security scanning or dead code detection (use code-security for that). Do NOT use for code quality review of specific PRs (use code-review for that). Do NOT use for running existing test suites (use code-preflight for that).

0
installs
▌
Postmortem · gridlock-nyc
Structured incident and bug analysis — timeline reconstruction, root cause identification, contributing factors, and prevention. Use when user says 'postmortem', 'what went wrong', 'why did this break', 'incident review', 'root cause analysis', 'RCA', 'failure analysis', 'what happened'. Do NOT use for session-level audits (use session-audit). Do NOT use for code quality review (use simplify).

0
installs
▌
Skill Eval · gridlock-nyc bundle
Benchmark how closely a skill followed its own instructions. Reads the skill's SKILL.md, extracts verifiable requirements, scores the skill's output from the current session. Returns compliance % with per-requirement pass/fail and cited evidence for every miss. Use when user says 'skill eval', 'did that skill work', 'how well did it follow instructions', 'benchmark skill', 'skill compliance', 'grade that skill', 'skill score', 'evaluate skill', 'skill audit', 'test skill'. Do NOT use for session-level audits (use session-audit for that). Do NOT use for building skills (use skill-builder for that). Do NOT use for behavioral evals of project code (use eval for that).

0
installs
▌
Chart Forge · gridlock-nyc
Generates professional inline SVG data visualizations — bar charts, donut charts, progress rings, stat cards, comparison tables, and sparklines — from user-supplied data. Output is clean, print-ready SVG embeddable in HTML documents, reports, and one-pagers without external dependencies. Trigger phrases: "chart forge", "make a chart", "generate a bar chart", "create a donut chart", "visualize this data", "add a chart to my report", "SVG chart", "data visualization", "make this visual", "progress ring", "stat card", "inline chart", "embed a chart", "chart from this data", "turn this into a chart". Do NOT use for: code analysis, architecture diagrams, flowcharts or system diagrams (use /plan-directions), data analysis or statistics (use exploratory-data-analysis), writing or editing document prose, generating images via external APIs, or any task that does not involve rendering data as SVG.

0
installs
▌
Code Review · gridlock-nyc
Staged code review — runs tests, extracts intent, dispatches 3 specialist agents (state-enumeration bug detection, project rule compliance, history + regression detection), then verifies findings with type-stratified confidence thresholds. Works on git diff (pre-commit) or PR diff. Use when user says 'code review', 'review this code', 'review changed code', 'review for regressions', 'check for bugs', 'is this good code', 'review what I just wrote'. Do NOT use for plan review (use /plan-review). Do NOT use for security/dead code scanning (use /code-security). Do NOT use for edge case auditing (use /code-edges). Do NOT use for pre-refactor impact analysis (use /code-blast-radius). Do NOT use for style cleanup or simplification (use /code-review).

0
installs
▌
Plan Review · gridlock-nyc bundle
Reviews plans from plan mode for real problems, then fixes them. Catches breaking bugs, security holes, over-engineering, and missing steps — but only when grounded in specific evidence. Returns CLEAN when a plan has no issues. Use when user says 'review this plan', 'check my plan', 'plan review', 'is this plan good', 'critique this plan', 'sanity check', 'poke holes in this', 'review my approach', or 'is this a bad idea'. Do NOT use for code review of written code (use simplify). Do NOT use for incident investigation (use postmortem). Do NOT use for generating alternative approaches (use directions).

0
installs
▌
Business Plan · gridlock-nyc bundle
Create investor-ready business plans by analyzing codebases and researching markets. Produces professional documents with product breakdown, market sizing, financial projections, go-to-market strategy, and implementation roadmaps. Use when user says 'business plan', 'strategic plan', 'market analysis', 'write a business plan', 'create a proposal', 'strategic assessment', 'consulting-style plan', 'plan for this project', 'business case', or 'go-to-market strategy'. Do NOT use for pitch decks, slide presentations, or investor update emails.

0
installs
▌
Code Security · gridlock-nyc
Scan the current project for security issues, dead code, deprecated patterns, forgotten debug code, secrets in config, and hidden oddities. Use when user says 'scan for issues', 'find security problems', 'codebase audit', 'treasures', 'what's wrong with this codebase', 'find dead code', or 'security scan'. Do NOT use for code review of specific PRs (use review for that). Do NOT use for refactoring suggestions (use code-review for that).

0
installs
▌
Plan Capacity · gridlock-nyc bundle
Assess whether a task is feasible for a Claude Code agent and produce a dispatch plan. Decomposes tasks into atomic units, classifies each using Anthropic's proven decision ladder, checks delegation specs, and outputs a concrete execution plan. Use when user says 'can an agent handle this', 'is this too much for one agent', 'agent capacity', 'scope this task', 'how should I break this down', 'task scoping', 'will this thrash', 'dispatch plan', 'how many agents', or 'is this a single agent job'. Runs after a plan is produced to validate the plan's task breakdown. Do NOT use for generating the plan itself (use plan mode for that). Do NOT use for reviewing plan quality (use plan-review for that). Do NOT use for choosing between architecture approaches (use plan-directions for that).

0
installs
▌
Session Audit · gridlock-nyc
Fix the system after a session — edit CLAUDE.md rules, rewrite confusing docs, propose hooks, delete misleading instructions, save memories. The output is changes, not a report. Use when user says 'audit', 'session audit', 'what went wrong', 'debrief', 'fix the system', 'why does this keep happening', 'session review'. Do NOT use for code review (use code-review) or production incidents (use postmortem).

0
installs
▌
Session Drift · gridlock-nyc
Detect scope drift in the current session. Compares all changes made against the original stated goal, flags tangential work, and identifies unfinished parts of the original task. Use when user says 'drift', 'am I on track', 'scope check', 'what was I doing', 'did I drift', 'focus check', or 'are we still on target'. Do NOT use for session auditing (use session-audit for that). Do NOT use for code review (use code-review for that).

0
installs
▌
Skill Builder · gridlock-nyc bundle
Build new Claude Code skills with best-practice enforcement from Anthropic's official skill guide. Handles use case definition, pattern selection, frontmatter generation, instruction writing, validation, and test case generation. Use when user says "build a skill", "create a skill", "make a new skill", "turn this into a skill", "I want a skill for", "skill for [X]", "make me a slash command", or "new skill that [does something]". Do NOT use for running evals or benchmarking existing skills (use skill-eval for that). Do NOT use for listing, deleting, or inspecting existing skills.

0
installs
▌
Code Preflight · gridlock-nyc
Run build, typecheck, lint, tests, and security checks for the current project. Auto-detects Python vs Node/Next.js.

0
installs
▌
Plan Directions · gridlock-nyc
Generate 2-4 concrete architecture approaches for a given task, with tradeoff analysis and a recommendation. Use when user says 'directions', 'what are my options', 'compare approaches', 'how should I build this', 'architecture options', 'design options', or 'explore approaches'. Do NOT use for code review (use code-review for that). Do NOT use for investigating bugs (use postmortem for that).

0
installs
▌
Research Search · gridlock-nyc
Rigorous evidence-based research — domain routing, evidence hierarchies, GRADE assessment, PRISMA screening, citation quality scoring. Use when user says 'investigate', 'research this', 'what does the evidence say', 'systematic review', 'literature review', 'deep dive', 'what do studies show'.

0
installs
▌
Code Blast Radius · gridlock-nyc
Maps all callers, importers, and dependents of a target function, class, module, or file before a refactor begins. Scores blast radius as low/medium/high and flags high-risk downstream effects so nothing breaks silently. Trigger phrases: "blast radius", "who calls this", "what depends on", "safe to rename", "safe to move", "what imports this", "refactor radar", "dependency map", "what breaks if I change", "callers of", "dependents of", "impact analysis before refactor", "check before refactoring". Do NOT use for: general code search, finding bugs, reviewing pull requests, auditing security, checking for dead code (use /code-security), detecting scope creep (use /session-drift), evaluating architecture options (use /plan-directions), or any task that is not specifically about pre-refactor impact analysis.

0
installs
▌
Research Critique · gridlock-nyc
Evaluate scientific claims, review manuscripts, assess evidence quality. Bias detection, GRADE assessment, logical fallacy identification, structured peer review reports. Use when user says 'review this paper', 'evaluate this claim', 'peer review', 'is this study valid', 'check the methodology', 'evidence quality', 'critical analysis'. Do NOT use for literature search (use research-search). Do NOT use for code review (use code-review).

0
installs

22 published skills