me-cedric/agent-compass

Central agentic standards, guidelines, skills & bootstrap tooling for AI coding agents (Claude, Codex, Copilot).

This repository also configures its own agents. See what agent-compass tells them →

3Stars on the repository
41Mods indexed here, across every type
4d agoLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

api-contract-sync

01

me-cedric/agent-compass

Skill Claude CodeCodex

Keep API code, OpenAPI, Bruno, Gherkin, mocks and client types in step. Use when an endpoint, DTO, payload, status code, query parameter or auth rule changes.

not rated 3 changed 10d ago A 43 tokens original MIT

me-cedric/agent-compass

Skill Claude Code

Recommend and justify the architecture of a new project from the needs of the client. Technology-neutral, research first, no unlabeled guess. Use for questions on which stack, framework or architecture to choose.

not rated 3 changed 10d ago A 44 tokens original MIT

delivery-digest

03

me-cedric/agent-compass

Skill Claude Code

Write the digest of what changed for the Product Owner or project manager, in plain language, with business impact and planning consequences and no code. Use for a stakeholder update, release notes or a delivery report.

not rated 3 changed 10d ago A 45 tokens original MIT

design-taste-skills

04

me-cedric/agent-compass

Skill Claude CodeCodex

Route frontend design direction work to the pinned taste skills. Use for a landing page, redesign, brand system or mockup-to-code task, not for a bug fix.

not rated 3 changed 10d ago A 40 tokens original MIT

impact-analysis

05

me-cedric/agent-compass

Skill Claude Code

Analyse what changed since a base git reference and write the developer action plan with technical impacts, blast radius and the project documents that are now stale. Use for the impact of a branch or diff before merge.

not rated 3 changed 10d ago A 44 tokens original MIT

me-cedric/agent-compass

Skill Claude CodeCodex

Route native Android and Apple work to the pinned vendor skills. Use for Kotlin, Compose, Gradle, Swift, SwiftUI, Xcode, an emulator or a simulator.

not rated 3 changed 10d ago A 40 tokens original MIT

project-memory

07

me-cedric/agent-compass

Skill Claude CodeCodex

Read and write durable project memory with projectmem. Use before non-trivial work, and after a decision, a failed attempt, a fix or a finding.

not rated 3 changed 10d ago A 35 tokens original MIT

me-cedric/agent-compass

Skill Claude CodeCodex

Route requests for terse, minimal or unambiguous output to the pinned style skills: caveman, ponytail, i-have-adhd, asd-ste100.

not rated 3 changed 9d ago A 0 tokens original MIT

e2e

09

me-cedric/agent-compass

Skill Claude CodeCodex

Run, write or heal end-to-end tests through ac run: Playwright and its test agents, Maestro, agent-device. Read the result and the reports.

not rated 3 changed 9d ago A 0 tokens original MIT

env

10

me-cedric/agent-compass

Skill Claude CodeCodex needs its repo

Keep env vars in one catalog, one template and the code. Check them with ac env check, share secrets with age, start apps with ac env run. Use when you add, rename, read or share an env var.

not rated 3 4d ago C 48 tokens original MIT

local-run

11

me-cedric/agent-compass

Skill Claude CodeCodex

Start the project stack locally with process-compose and docker compose. Use to run, restart or debug the local apps, or to write process-compose.yaml.

not rated 3 4d ago A 33 tokens original MIT

vuln-triage

12

me-cedric/agent-compass

Skill Claude CodeCodex

Triage security findings from osv-scanner, Checkmarx and SonarQube. Merge the SARIF files, rank by CVSS, then fix or suppress with a reason. Use when a scan, a CVE alert or a dependency audit needs action.

not rated 3 4d ago A 60 tokens original MIT

adr

13

me-cedric/agent-compass

Skill Claude Code

Write a MADR 4 decision record from a meeting transcript, call notes or a decision made in chat. Keeps the rejected options and the reasons. Use when the user asks to record or capture a decision.

not rated 3 4d ago A 44 tokens original MIT

ci-triage

14

me-cedric/agent-compass

Skill Claude CodeCodex

Triage a red CI job: read the log, classify it as flaky, environment or real, reproduce it, fix small and rerun. Use when a pipeline or check is red.

not rated 3 changed 9d ago A 0 tokens original MIT

compass

15

me-cedric/agent-compass

Skill Claude Code

Set up, update, migrate, extend and promote agent-compass in a project with the ac CLI. Use when the user asks to install, sync, upgrade or extend compass, or to move a project skill into compass.

not rated 3 4d ago A 48 tokens original MIT

debug

16

me-cedric/agent-compass

Skill Claude CodeCodex

Build a failing loop before you diagnose a hard bug, flaky test, regression or slowdown, with read-only Grafana, ArgoCD or Kubernetes evidence. Use when a report says only "broken".

not rated 3 changed 9d ago A 42 tokens original MIT

grill

17

me-cedric/agent-compass

Skill Claude Code

Interview the user about a plan until each open question has an answer or an owner. Writes one RAID file per open question and one ADR per decision. Use to stress-test a plan or to fill the RAID register.

not rated 3 4d ago A 46 tokens original MIT

harness-loop

18

me-cedric/agent-compass

Skill Claude Code

Run a large goal as a loop with one check, a progress file on disk and a turn cap. Uses the deliver workflow for multi-file work. Use when the user asks you to keep going until a big goal is finished.

not rated 3 4d ago A 50 tokens original MIT

remaining-work

19

me-cedric/agent-compass

Skill Claude Code

Audit what the code delivers against the specs, then write the remaining-work backlog and the completion plan (CP view). Use when asked how far along the project is, what is left, or for a roadmap.

not rated 3 4d ago A 45 tokens original MIT

review

20

me-cedric/agent-compass

Skill Claude CodeCodex

Review a change, a branch or PR/MR N on spec and standards. For PR/MR N, show a recap, then post the summary, inline comments and verdict. Use before done.

not rated 3 changed 9d ago A 42 tokens original MIT

spec-flow

21

me-cedric/agent-compass

Skill Claude CodeCodex

Write, validate and archive OpenSpec specs (konvoy-delivery schema); map screens, convert docs, draft specs from code, triage drift. Use before code for a feature or unclear task.

not rated 3 changed 9d ago A 43 tokens original MIT

task-flow

22

me-cedric/agent-compass

Skill Claude CodeCodex

Draft, split, assign and move tickets on Jira, GitLab or GitHub issues, from specs. Use when the user names a ticket or asks to write, split or assign tickets.

not rated 3 changed 9d ago A 41 tokens original MIT

ui-conformance

23

me-cedric/agent-compass

Skill Claude CodeCodex

Prove a screen matches its design: tokens, computed style, structure, then a Playwright baseline. Use when you change UI or build from Figma.

not rated 3 changed 9d ago A 0 tokens original MIT

harness-improve

24

me-cedric/agent-compass

Skill Claude CodeCodex

Run ac reflect on recent sessions, fix repeated failures in project-owned harness files, and record one lesson. Use when an error or a blocked action repeats.

not rated 3 4d ago A 36 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: