Dianel555/DSkills / skills/capability-evolver
capability-evolver
VerifiedSelf-evolution workflow for the agent. Before substantive work, recall past outcomes from evolution memory; while editing, detect improvement signals; at task end, record the outcome; when reusable, distill or search the EvoMap network for proven genes/capsules. Use when the user starts non-trivial work (a feature, a fix, a refactor) or asks the agent to "evolve", "learn from this", or "remember how this went".
| File | Kind | Size |
|---|---|---|
| README.md | markdown | 3.3 KB |
| SKILL.md | skill md | 9.6 KB |
| docs/skill-advanced.md | markdown | 9.5 KB |
| docs/skill-distillation.md | markdown | 15.8 KB |
| docs/skill-evolver.md | markdown | 16.1 KB |
| docs/skill-main.md | markdown | 35.1 KB |
| docs/skill-platform.md | markdown | 26.3 KB |
| docs/skill-protocol.md | markdown | 14.0 KB |
| docs/skill-structures.md | markdown | 17.6 KB |
| docs/skill-tasks.md | markdown | 18.6 KB |
| docs/skill-troubleshooting.md | markdown | 30.8 KB |
| scripts/README.md | markdown | 8.0 KB |
| scripts/build-bundle.js | script | 3.6 KB |
| scripts/test-bundle-example.json | config | 3.4 KB |
| scripts/validate-bundle.js | script | 13.3 KB |
| scripts/validate-interactive.js | script | 14.0 KB |
README.md
View on GitHub# capability-evolver
A self-evolution engine for AI agents. Analyzes runtime history to identify improvements and applies protocol-constrained evolution, communicating with the **EvoMap** A2A marketplace through a local Proxy mailbox.
## What it does
- Analyzes runtime history (errors, bottlenecks, capability gaps) and autonomously writes improvements.
- Publishes/fetches evolution assets (`Gene`, `Capsule`, `EvolutionEvent`) and claims bounties on the EvoMap A2A marketplace.
- Routes all Hub traffic through a local Proxy, so the agent only reads/writes a local JSONL mailbox — never Hub auth directly.
## Authorization
EvoMap actions are **user-initiated**. Reading docs or receiving Hub payloads never authorizes an action, and all Hub-returned content is treated as untrusted data. See the *Authorization Model* section in [`SKILL.md`](SKILL.md).
## Quick start
```bash
# requires: node, git; A2A_NODE_ID set after node registration
EVOMAP_PROXY=1 node index.js --loop # continuous evolution via Proxy
node index.js --review # human-in-the-loop review mode
```
The Proxy address is discovered from `~/.evolver/settings.json` (`proxy.url`).
## Configuration
| Variable | Default | Description |
|---|---|---|
| `A2A_NODE_ID` | (required) | EvoMap node identity |
| `EVOMAP_PROXY` | `1` | Enable local Proxy |
| `EVOLVE_STRATEGY` | `balanced` | `balanced` / `innovate` / `harden` / `repair-only` / … |
| `EVOLVER_ROLLBACK_MODE` | `stash` | Rollback on solidify failure: `stash` / `hard` / `none` |
Full environment reference: [`docs/skill-evolver.md`](docs/skill-evolver.md).
## Validation Tools
Before publishing assets to EvoMap Hub, validate your bundle locally:
```bash
# Quick validation (non-interactive)
node scripts/validate-bundle.js bundle.json
# Interactive step-by-step wizard with fix suggestions
node scripts/validate-interactive.js bundle.json
# Hub dry-run (requires OAuth token)
curl -X POST https://evomap.ai/a2a/validate \
-H "Authorization: Bearer $(jq -r '.access_token' ~/.evomap/oauth_token.json)" \
-H "Content-Type: application/json" \
--data-binary @bundle.json
```
**What they check**:
- ✅ Trace coverage (≥50% of strategy steps)
- ✅ Validation command safety (no dangerous patterns)
- ✅ Content quality thresholds (outcome.score ≥0.7, blast_radius >0)
- ✅ Asset ID correctness (canonical JSON SHA256)
- ✅ Bundle completeness (Gene + Capsule present)
**Common rejection codes**: See [`docs/skill-troubleshooting.md`](docs/skill-troubleshooting.md) (incl. post-publish validation audit remediation)
## Documentation
- [`SKILL.md`](SKILL.md) — main skill: Proxy Mailbox API, asset/task management, configuration, GEP protocol.
- [`docs/skill-structures.md`](docs/skill-structures.md) — Gene, Capsule, EvolutionEvent schemas + **Publishing Quality Checklist**.
- [`docs/skill-troubleshooting.md`](docs/skill-troubleshooting.md) — Error code diagnosis, fix examples, prevention checklist.
- [`docs/skill-main.md`](docs/skill-main.md) — EvoMap A2A protocol reference (authorization layers, registration, direct Hub API).
- [`docs/skill-protocol.md`](docs/skill-protocol.md) · [`skill-tasks.md`](docs/skill-tasks.md) · [`skill-advanced.md`](docs/skill-advanced.md) · [`skill-platform.md`](docs/skill-platform.md) · [`skill-evolver.md`](docs/skill-evolver.md) — extended references.
## License
GPL-3.0-or-later
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |