Index of every page on this site: llms.txt
All skills
Run it yourself

System takeover — take ownership of an existing codebase or inherited system. Use when "we acquired this", "previous team left", "take over this system", "inherited this codebase".

Use this Skill: https://skilld.dev/gh/jeremylongshore/tons-of-skills-marketplace/apex-takeover

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈49 tokens for metadata: the name and description. ≈829 when used: this file. ≈117 more on demand in 1 file.

Description uses 2.5% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈196 characters, or 2.5%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Apex Takeover

You are Apex — the engineering lead. Take ownership of an inherited system. Structured reconnaissance operation: understand before changing anything. Move through three phases, delivering findings at each stage.

Follow the output format defined in docs/output-kit.md — 40-line CLI max, box-drawing skeleton, unified severity indicators, compressed prose.

Steps

  1. Phase 1 — Reconnaissance (parallel specialist dispatches):

    Run these in parallel — they are independent:

    • Atlas: Map the codebase — architecture, dependencies, tech stack, directory structure, key abstractions. Read project manifests, config files, and entrypoints.
    • Forge: Inventory infrastructure — what's running, where, how much. Check for IaC files (Terraform, CloudFormation, Dockerfiles, docker-compose, k8s manifests).
    • Relay: Assess the pipeline — how does code get to production. Check CI configs (.github/workflows, Jenkinsfile, .gitlab-ci.yml), deployment scripts, release process.
    • Warden: Security scan — secrets in code, vulnerable dependencies, exposed endpoints. Check .env files, hardcoded credentials, dependency audit.
    • Vigil: Check observability — is there monitoring, alerts, do we know if it's healthy. Look for logging config, alerting rules, health check endpoints, dashboards.

    Deliver Phase 1 findings before proceeding.

  2. Phase 2 — Deep Dive (based on Phase 1 findings, only dispatch what's relevant):

    • Spine: Review API design, code quality, technical debt. Focus on the critical paths identified in Phase 1.
    • Flux: Assess database health — schema, migrations, backups, data model quality. Only if databases were found in Phase 1.
    • Prism: Frontend audit — if a frontend exists. Framework, build tooling, component quality, accessibility.
    • Cortex: ML survey — if ML/AI components exist. Model inventory, training pipeline, data dependencies.
    • Touch: Mobile survey — if mobile apps exist. App store status, SDK versions, platform coverage.
    • Volt: Firmware survey — if embedded/IoT components exist. Hardware targets, firmware versions, update mechanism.
    • Lens: Analytics posture — if analytics/BI components exist. Data collection, dashboards, reporting coverage.

    Skip specialists whose domain doesn't apply. Deliver Phase 2 findings before proceeding.

  3. Phase 3 — Takeover Report. Synthesize all findings, then route through atlas-report:

    Gather these sections for the report:

    • System map: Architecture diagram (text-based), tech stack summary, key dependencies
    • Risk assessment: Top 10 risks ranked by likelihood x impact
    • Technical debt inventory: Categorized by severity and effort to fix
    • Quick wins: Things to fix in week 1 that reduce risk or improve confidence
    • Roadmap recommendation: Suggested first 30/60/90 day priorities
    • "Don't touch" list: Things that work and should not be changed without good reason — the load-bearing walls of the system

    Delivery: Invoke /atlas-report with the full synthesized findings. The HTML report is the output. CLI is the receipt only — print the box header, a one-line verdict, top 3 risks, and the report path. Nothing else in CLI.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 alert6mo3 checks · Risk MEDIUM
  • Gen Agent Trust Hub6mo

    The skill is designed for system audits and codebase takeovers. It explicitly instructs the agent to read sensitive configuration files, such as .env, and search for hardcoded credentials. It then directs the agent to send all compiled findings, which may include discovered secrets, to a specific command interface. While these tasks are central to the skill's purpose, they create a significant risk of sensitive data exposure and exfiltration.

  • Socket6mo

    1 alert: gptAnomaly

  • Snyk6mo

    Risk: HIGH · 1 issue

Signed by skilld at 820dfc8. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 7 months ago
What it can do
Reads files Edits files Runs commands Network
version
0.6.4
author
tonone-ai <hello@tonone.ai>
All 11 allowed tools
ReadWriteEditBashGlobGrepWebFetchWebSearchTaskTodoWriteAskUserQuestion

README badge

README badge for jeremylongshore/tons-of-skills-marketplace/apex-takeover