Practices for building software with coding agents, each with a test you check on your repo.
Scored across 4 tools
Each tool has a distinct role: overview onboards, find_elements searches by name/role/group/row, get_element returns one full record, and repository_checklist applies tests to a project. find_elements and get_element form a clear search/detail pair, leaving little room for misselection.
find_elements and get_element follow a clean verb_noun pattern, but overview and repository_checklist break it as noun-only names. The mix is still readable and unambiguous, but the convention is not uniform across the set.
Four tools is well-matched to a reference/knowledge-base domain: onboarding, search, detail retrieval, and application checklist. Nothing feels redundant or missing at the count level, though it sits near the thin end.
The surface covers the full retrieval lifecycle (overview, search, detail, practical application). A dedicated way to list/enumerate all elements or cross-compare multiple elements is absent, but find_elements with broad queries likely covers enumeration.
Find elements of the table by words in their name, problem or test, by role (dev, lead, plat, qa, ...), by group (security, verification, ...) or by row (task, change, system, organisation). Returns one line per element.
Inputs: row, role, group, query.
One element in full: problem, observable test with its clauses and evidence kinds, when it applies, when it adds work without helping, activities, relations, page link. By ID (ae-006), symbol (Sx) or name.
Inputs: ref (required).
The Agentic Engineering Periodic System: what it is, how to read it (rows, columns, judging and reading rules), the roles. Call this first.
Inputs: no named parameters.
What a repository can show for each test, clause by clause, what it cannot, and the judging rules. Use it to check the current project; read the files yourself. Optional: element IDs, a role or a group to narrow it.
Inputs: ids, role, group.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Real coding-agent sessions: what worked, where agents failed and what the human did about it.
Writes adversarial test suites for AI-built code. Your agent's test engineer.
Visual testing for coding agents: your agent sees every UI change it makes and triages each diff.
Change-aware CI validation and affected-test guidance for coding agents.
Add one secure layer between your agents and this server.