[SyedFahad7]

Adds steps 5 to 7 to Setup for AI agents on for-agents.mdx (served as /for-agents.md):

5. Ask before adding a short "Graphify before grep" rule to the project's existing AGENTS.md / CLAUDE.md / .cursor/rules/graphify.mdc; create only if none exists.
6. Run one real query on a returned repo and show the answer.
7. One-line handoff saying the agent will query the graph from now on, linking the cookbook.

Step 4's report also names the rule file; intro says steps 1 to 7. Cookbook rule aligned: grep for exact strings; on failure say so, then fall back to grep (was "do not silently fall back"). Tool names match the live hosted MCP tool list (checked 30 Oct). Homepage hero prompt unchanged. Not for merge without Fahad's OK.

[mintlify]

Preview deployment for your docs. Learn more about Mintlify Previews.

Project	Status	Preview	Updated
graphify-labs-inc	🟢 Ready	View Preview	Oct 30, 2026, 1:30 PM

💡 Tip: Enable Automations to automatically generate PRs for you.

[graphify-labs]

Graphify reviewed this change.

No issues found by static checks — no coupling regressions or blocking issues in the code graph. That is not the same as safe to merge: see what was not checked.

Not checked: tests were not run; no formal proof of the changed code.

---

Graphify review — findings

Extends the agent setup flow in for-agents.mdx from four steps to seven: after connecting, the agent offers to append a short Graphify rule to the project's instruction file (AGENTS.md, CLAUDE.md, or a Cursor rule), runs one real query against a repo from list_repositories, and closes with a one-line "setup done" message that links the cookbook. The rule only gets written if the user says yes, and it is appended to an existing file rather than replacing it. The Cursor .mdc path and the step 4 report, which names a file that step 5 only writes afterward, are where the ordering and file handling can go wrong. Also loosens the cookbook's instruction snippet so agents use grep for exact strings and fall back to it, with a stated failure, when a Graphify call fails, instead of the old "never silently fall back" rule.

No blocking issues surfaced. 2 lower-confidence candidates did not survive cross-model review.

Analysis details — impact, health, verification

Impact & health

Graphify review

Impact — 23 functions depend on the 23 functions this change touches.

Health — grade A; no new coupling hotspots.

Verification — 23 functions in the blast radius were not formally verified this run (proofs are advisory here).

Gate & verification

graphify gate

PASS — objectively clean (no health regressions, tests not run — proofs not run this pass (advisory)). Grounded, not self-assessed.

Advisory (not blocking):

• verification_scope: 23 function(s) in the blast radius were not formally verified this run

Test selection

Test selection

0 test file(s) selected via static blast radius.

Escalated to a full run for safety — the selection is not trustworthy on its own (see below). CI should run the whole suite.

non-code file(s) changed (cookbooks/make-agents-use-the-graph.mdx, for-agents.mdx) → running the full suite for safety (a code graph can't see config/fixture/data deps)

changed code file(s) with no mapped test (cookbooks/make-agents-use-the-graph.mdx, for-agents.mdx) — a coverage gap or a missing link — running the full suite rather than only the selected tests

Selection is safe under the controlled-regression assumption; always-run tests + a periodic full run are the backstops. Advisory — it never changes the check verdict.