Tarekkharsa
UserLocal control plane for AI coding agents — trust-gate, firewall, and audit every MCP tool call. One manifest for servers, skills, instructions, and settings across Claude Code, Codex, Cursor, Gemini CLI + 9 more.
Categories
Indexed Skills (18)
mine-skills
Mine your past agent sessions (Claude Code, Codex) for recurring failures and re-explained procedures, then distill the best candidates into reusable skills that land in the central library via `agentstack lib add`. Local and read-only until you approve a draft.
orchestrate-workflow
Wire a generate-review-fix multi-agent workflow (one implementer, N adversarial reviewers, a fixer) from agentstack profiles — models per role, skills per role, secrets injected at start — and run it in your executor of choice (sandcastle, Claude Code workflows, or plain Docker). agentstack provisions and governs the agents; it never runs the loop.
using-agentstack
Operate agentstack — manage a project's MCP servers, skills, profiles, and secrets from its manifest; work through the trust-gated runtime gateway; interpret policy denials; activate/deactivate capability sets; load skills from the central library; propose new capabilities safely.
finding-skills
Find and install agent skills through agentstack — search the library/catalog/registry, pull from any skills repo (owner/repo, git URLs, local dirs), and judge quality by scan verdicts, pins, and provenance instead of install counts. Use when the user asks "how do I X", "is there a skill for X", or wants a capability added.
propose-workflow
Propose a reviewable multi-agent workflow as a blueprint — pick and name the pattern (map-reduce, pipeline, tournament, loop-until-dry, dag), emit its shape as an agentstack-blueprint JSON block, and WAIT for the user to approve / reject / edit before authoring and running it via agentstack workflow run.
adversarial-review
Review a diff as an adversary — assume the code is wrong, work from the diff alone, and try to construct concrete failures. The reviewer-role procedure for multi-agent workflows (one implements, N review), distilled from the Bun-in-Rust port.
cloudflare_ship_worker
Ship a Cloudflare Worker correctly — configure wrangler, set bindings and secrets, run a dry-run, deploy, and verify.
cloudflare_wrangler_cheatsheet
Quick reference mapping Cloudflare product surfaces (Workers, Pages, KV, R2, D1, DNS) to the wrangler CLI commands teams use day to day.
posthog_query
Turn a product or analytics question into the right PostHog query — trends, funnels, or retention — with the correct events, breakdowns, and date range.
run-codex
Delegate clear-spec, bulk, or mechanical work — migrations, codemods, large refactors, log digs, and independent code review — to gpt-5.5 through the Codex CLI, with quota-aware fallback to Claude models.
sync-library
Keep your agentstack central library (~/.agentstack/lib) consistent across machines by versioning it as a git repo — commit the index and path-source skill bodies, exclude the content store cache and any resolved secrets.
analyze-usage
Report on your own agent usage — parse Claude Code session transcripts for token burn, and cross-check installed agentstack skills/servers against what was actually used to flag dead weight. Local and read-only.
linear_breakdown
Break a large Linear issue into small, shippable tickets with clear scope, estimates, and dependencies.
posthog_experiment
Set up a sound PostHog A/B experiment or feature flag rollout — hypothesis, primary metric, exposure, and a sample-size check before launch.
pr-triage
Triage a stack of open pull requests into review / merge / close buckets with a clear next action for each.
route-by-cost
Pick the right model for each job — route bulk and mechanical work to cheap models, exploration to fast mid models, and reserve the strongest model for judgment, taste, and final review. Escalate when the output misses the bar.
api-conventions
House API conventions for the storefront — REST resource shapes, the shared error envelope, and pagination rules.
sql-review
Review SQL for correctness, performance, and safety before it ships.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.