Works with the AI Agents you already use
SysEdge
Map features, audit use-case quality, track test coverage, enforce architecture standard, drive agent work on defects and enhancements, shared context across the team. Agents start with a briefing not a codebase crawl.
$29
SysEdge
Example session with this skill installed
Setup Sysedge on this project.
- Read your context and instructions
- Compiled the sysedge
Skill(sysedge)
Successfully loaded skill
Step 1 — Initialise
Ran 2 shell commands
Empty graph. Steps 2–4 — Detect project structure, map sources and tests
Searched for 1 pattern, ran 1 shell command
Go + TypeScript/Preact project. 5 Go test files, 12 Playwright specs. Layout matches what I already know well — proposing the seed directly.
Step 5 — Proposed modules and features
┌─────────────┬──────────┬────────────────┬─────────────────────────────┬─────────────────────────────────────────┐
│ Module │ Instance │ Domain │ Code path │ Description │
├─────────────┼──────────┼────────────────┼─────────────────────────────┼─────────────────────────────────────────┤
│ MOD-server │ kbedge │ Backend │ cmd/kbedge/, │ Streamable HTTP + stdio MCP wiring, │
│ │ │ │ internal/mcpserver/ │ panel routes │
├─────────────┼──────────┼────────────────┼─────────────────────────────┼─────────────────────────────────────────┤
│ MOD-auth │ kbedge │ Backend │ internal/auth/ │ dev / token / Entra modes, Claims, │
│ │ │ │ │ PanelTokenStore │
├─────────────┼──────────┼────────────────┼─────────────────────────────┼─────────────────────────────────────────┤
│ MOD-config │ kbedge │ Backend │ internal/config/ │ Env-driven config, timezone │
├─────────────┼──────────┼────────────────┼─────────────────────────────┼─────────────────────────────────────────┤
│ MOD-store │ kbedge │ Backend │ internal/store/, │ pgx pool, sqlc queries, WithTx │
│ │ │ │ migrations/ │ │
├─────────────┼──────────┼────────────────┼─────────────────────────────┼─────────────────────────────────────────┤
│ MOD-service │ kbedge │ Backend │ internal/service/ │ Business logic, view assembly, audit │
│ │ │ │ │ calls │
├─────────────┼──────────┼────────────────┼─────────────────────────────┼─────────────────────────────────────────┤
│ MOD-views │ kbedge │ Backend │ internal/views/ │ View structs → generates types.ts │
├─────────────┼──────────┼────────────────┼─────────────────────────────┼─────────────────────────────────────────┤
│ MOD-rules │ kbedge │ Backend │ internal/rules/ │ Attention + priority rules (BR-xx) │
├─────────────┼──────────┼────────────────┼─────────────────────────────┼─────────────────────────────────────────┤
│ MOD-audit │ kbedge │ Backend │ internal/audit/ │ Immutable audit writer (same-tx) │
├─────────────┼──────────┼────────────────┼─────────────────────────────┼─────────────────────────────────────────┤
│ MOD-spa │ kbedge │ Frontend │ web/src/ │ Preact SPA: Board, MyWork, Attention, │
│ │ │ │ │ TaskDetail │
├─────────────┼──────────┼────────────────┼─────────────────────────────┼─────────────────────────────────────────┤
│ MOD-infra │ kbedge │ Infrastructure │ deploy/, seed/, scripts/ │ Docker Compose, migrations, seed loader │
└─────────────┴──────────┴────────────────┴─────────────────────────────┴─────────────────────────────────────────┘
┌──────────────┬─────────┬────────────────────────────────────────────────────────────────────┐
│ Feature │ Module │ Description │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SRV-001 │ server │ Streamable HTTP MCP endpoint (POST /mcp) │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SRV-002 │ server │ Panel token mint + redirect (/app → /panel/{token}) │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SRV-003 │ server │ Panel view API (/api/panel/{token}/view) │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SRV-004 │ server │ Health check (/healthz) │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-AUTH-001 │ auth │ Dev mode auto-claims │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-AUTH-002 │ auth │ Static bearer token validation │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-AUTH-003 │ auth │ Entra (Azure AD) JWT validation │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-AUTH-004 │ auth │ In-memory panel token store (1h TTL) │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-CFG-001 │ config │ KBEDGE_* env-driven configuration │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-STR-001 │ store │ sqlc-generated typed queries + pgx pool │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-STR-002 │ store │ Transactional write helper (WithTx) │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-STR-003 │ store │ Full-text task search │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SVC-001 │ service │ Board and column management │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SVC-002 │ service │ Task CRUD (create/update/close/assign) │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SVC-003 │ service │ Task column movement │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SVC-004 │ service │ Comments and activity thread │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SVC-005 │ service │ Board member management │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SVC-006 │ service │ View assembly │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-VW-001..4 │ views │ Board, TaskDetail, MyWork, Attention view structs │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-RUL-001 │ rules │ Priority scoring (BR-xx) │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-RUL-002 │ rules │ Overdue and stalled detection │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-RUL-003 │ rules │ WIP limit advisory │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-AUD-001 │ audit │ Immutable audit event append │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-AUD-002 │ audit │ Audit log view for administrators │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-SPA-001..6 │ spa │ Board, MyWork, Attention, TaskDetail, BoardPicker, Standalone mode │
├──────────────┼─────────┼────────────────────────────────────────────────────────────────────┤
│ F-INF-001..3 │ infra │ Docker Compose, migrations, seed loader │
└──────────────┴─────────┴────────────────────────────────────────────────────────────────────┘
Does this look right? Confirm and I'll write the seed JSON, load it, scan tests and wire up the links.
✻ Cogitated for 29s · done 4:58 PM
❯ yes
Ran 1 shell command
Seed file is already there. Seeding, scanning and linking:
Ran 2 shell commands
53 nodes seeded, 132 tests registered. Now running suggest-test-links and applying the high-confidence links:
Ran 1 shell command
No auto-matches (feature names don't contain UC/US IDs). Applying all known links in one pass:
Ran 2 shell commands
SysEdge is set up. 10 modules · 36 features · 132 tests · 15/36 (41%) covered — backed up to data/sys-backup-2026-11-08T16-58-53.json.
Notable gaps to address next
- F-RUL-002/003 — overdue/WIP rules are tested in uc-08 but not yet linked to MOD-rules
- F-SPA-004/005/006 — task detail drill-down, board picker, standalone mode (new features, no E2E spec yet)
- F-AUD-001 — audit write has no dedicated test
- MOD-views — view structs have no tests at all
Use SE briefing --instance kbedge at the start of each session and SE backup at the end.
✻ Crunched for 38s · done 4:58 PM
Connects securely to your tools. The creator never sees your data.
About this skill
AI agents lose context between sessions. They re-read the codebase, miss cross-module dependencies, and repeat work already done. SysEdge fixes this by maintaining a local requirements traceability graph that connects user stories, use cases, and features to code symbols, API endpoints, and test suites — giving every agent a shared ground truth it can query in seconds.
What it does
- Feature mapping connects high-level capabilities to specific source files, symbols, and API endpoints across Go, TypeScript, Python, Java, and C#.
- Use-case quality audits check that every feature is backed by well-formed use cases with acceptance criteria before work begins.
- V-model test coverage tracks coverage across four tiers — unit, integration, use-case, and E2E — and surfaces gaps by module with language-aware hints.
- Architecture standards are loaded into the graph and enforced during quality reviews, so agents flag violations before they reach review.
- Defect and enhancement lifecycle enforces a log-before-fix workflow and claim/close handoff so concurrent agents don't collide on the same work.
- Session briefing generates a context-aware summary of open work, coverage gaps, and blockers — agents start informed, not from scratch.
How it works
- Bootstrap — SE init, scan the directory structure, seed a JSON map of modules and features.
- Scan — SE scan-tests, SE scan-code, and SE suggest-test-links register tests and symbols and propose feature links.
- Work — SE start-enhancement and SE link-defect gate every session; SE briefing opens each one.
- Analyse — SE test-gaps, SE coverage-report, and SE impact keep implementation honest throughout.
Use cases
- Adding traceability to a legacy codebase without rewriting documentation.
- Running parallel agent sessions on the same repo without conflicts.
- Enforcing defect tracking and verification before any code change.
- Identifying features with no test coverage before a release.
Known limitations Does not perform Git operations — commits and pushes remain with the user. Requires Python 3.
Requires Python 3. No automatic Git operations. Local SQLite storage only.
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 any payment.
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
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 4 days ago
Requires
Permissions
Allowed Hosts