Reuse Docbot application specs as Explorbot context

[DenysKuchma]

Summary

Adds support for reusing documentation generated by Docbot as additional context in Explorbot.

Docbot page files now include a versioned application-spec contract. Explorbot can load the generated bundle from configuration or through the --spec CLI option, match documentation to the current URL, and provide it to Researcher, Planner, Navigator, Tester, and other agents.

Changes

• add applicationSpec configuration option
• add --spec <path> to web CLI commands
• introduce a validated, versioned application-spec format
• load only documentation relevant to the current page
• keep live HTML, ARIA, and screenshots as the primary source of truth
• treat User Might capabilities as unverified context
• separate application-spec context from regular knowledge
• support Docbot bundle directories and index.md paths
• tolerate SPA-added URL fragments during navigation
• document the Docbot-to-Explorbot workflow

[github-actions]

Explorbot Self-Regression

Commit 3433fc9 · run

Scenario	Result	Attempts	Duration
basic (native)	PASS	1/3	10m
experience: control	OK — failed as expected	1/1	1m
experience: seeded	PASS	1/3	1m

Attempt details

• basic (native) attempt 1 — PASS: login evidence: PASS (post-login plan=true, post-login research=true); research: PASS (files=6, wellFormed=true, keywords=7/3); scenarios: PASS (tests=6/5, features=4/3); tests passed: PASS (5 passed, 0 failed (reporter: 5 passed, 0 failed))
• experience: control attempt 1 — PASS: control: OK — failed as expected (0 passed, 1 failed)
• experience: seeded attempt 1 — PASS: seeded: PASS (1 passed, 0 failed)

---

Session analysis — basic (native):

Session Analysis

The Issues list feature was explored across 5 test scenarios covering create, search, and filter operations. All core flows executed successfully with the new issue appearing correctly in the list, search returning matching results, and label/status filters working as expected.

Coverage

• Pages: /issues, /issues?new=1, Add deep interactive exploration mode for researcher #7
• Features: Create issue with title/priority/label/status, Search by keyword, Filter by label, Filter by status, View issue details

What works

• Create issue — ET-1, ET-5 — new issues appear in table with correct status, priority, and labels
• Search — ET-2 — returns only matching issues, correct row visible
• Filter by label — ET-3 — shows only bug-labeled issues
• Filter by status — ET-4 — shows only open issues

Execution Issues

• ET-1 — first form fill failed, retried with different selectors then passed
• ET-5 — status change button click failed, dropdown reopen succeeded; initial verification failed but subsequent assertions passed

[DavertMik]

We also need document what application SPEC is into docs/ so non-docbot specs could also be accepted