- Home
- Skills
- Agents & Orchestration
- Aegis Skill Tree
Aegis Skill Tree
Plan capability growth as an evidence-backed skill tree with guarded upgrades, persistent history, and reversible rollback records.
Free
Works with the AI Agents you already use
Aegis Skill Tree
Example session with this skill installed
Inspect the supplied skill-tree snapshot. Identify the next useful upgrade, check its dependencies and evidence, propose a revision with acceptance tests and a rollback target, and do not apply any change.
- Read your context and instructions
- Compiled the aegis tree
Skill Tree Review
Observed state
- Active revision: rev-004
- The tree contains governance, continuity, and memory branches.
- governance.contracts is proficient and governance.skill-tree is active.
- memory.rollback is available and depends on governance.skill-tree.
- The snapshot contains an applied upgrade, a rolled-back synthetic upgrade, and a separate rollback revision. The earlier revision remains in the chain.
Recommended next step
Prepare an evidence-backed proposal to activate memory.rollback. First verify that the rollback target and dependency checks are available. Do not change the active tree yet.
Proposal
- Parent revision: rev-004
- Proposed kind: upgrade
- Risk: medium because the capability changes continuity behavior.
- Approval: required before application.
- Acceptance tests: validate the snapshot, verify the dependency graph, run the rollback-target check, and record the actual results.
- Rollback target: rev-004 or another host-confirmed known-good revision.
Boundaries
This is a plan, not an applied change. No file, network, or host state was changed. If verification fails, keep the failed revision and its evidence, restore only through the host rollback API, and append a new rollback event instead of deleting history.
Connects securely to your tools. The creator never sees your data.
About this skill
Aegis Skill Tree helps an AI agent map capabilities as a dependency tree, identify evidence-backed gaps, propose the next upgrade, and prepare a reversible revision or rollback record. It keeps current state separate from append-only history so a rollback changes the active revision without erasing the failed revision or its evidence. The package includes a portable snapshot contract, a worked rollback review, and an optional Python 3.9+ standard-library checker.
This is a standalone, instruction-led skill for SKILL.md-compatible agents. It is not the Aegis Stable runtime, does not train a model, and does not provide a sandbox, kill switch, approval service, budget enforcement, or hidden self-editing engine. Host integrations must expose those controls and provide direct evidence before the agent claims they ran.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
This package is advisory and host-dependent. It does not implement the Aegis Stable runtime or provide enforcement. The bundled example is synthetic, and a valid snapshot is not proof that a host applied or verified a change. Full revision storage, approval, sandboxing, and rollback must be supplied and evidenced by the host.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 16 days ago
- Free with an Agensi account
Nothing to install. It runs in your AI agent.
- Optional
Python 3.9+
Python only runs the optional local checker.
Permissions
File Scopes
Read-only by default. The optional checker reads JSON and prints validation results. No network, credential, or environment-variable access is required. Any host mutation or rollback must be separately exposed and explicitly approved.