release: 5.2.2 — front-door UX flow fixes (first command, gate docs, economy config)#16
Merged
Conversation
UX-P0-1: README led with /atlas which a fresh plugin install doesn't provide
(commands are namespaced /prd:*). Add brand 'atlas' entrypoint skill (/prd:atlas
-> go) and fix README first-run.
UX-P0-2: rewrite setup/discover/generate/handoff entry gates — they documented a
STOP then told the agent to ignore it ('proceed past... see morning brief'). Now
explain check_gate is an EXIT gate; first-entry false is expected. Removed leaked
internal refs (morning brief, Mum dogfood feedback).
UX-P0-3: load_fleet_config now honors token_economy from .atlas-ai/config/atlas.json
(what /customise-workflow writes); fleet.json still wins. Schema doc updated.
Version hygiene: bump __init__.py (5.2.1 missed the source-of-truth).
334 tests green. See docs/audit/UX-FLOW-AUDIT.md.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Front-door UX flow fixes → 5.2.2
5.2.1 hardened the engine; a follow-up UX-flow audit (
docs/audit/UX-FLOW-AUDIT.md) found the journey still broke before a new user ever reached those fixes. All three P0 flow-breakers were verified firsthand and are fixed here.The 3 flow-breakers (verified + fixed)
/atlas, but a fresh/plugin install prdprovides nocommands/and noatlasskill (plugin commands are namespaced/prd:*). → Added a brandatlasentrypoint skill (/prd:atlas, a thin alias dispatching to thegoorchestrator) and fixed the README first-run to/prd:atlas//prd:go/ natural language.setup/discover/generate/handoffopened with "if the gate fails, stop" then immediately "it WILL fail on first entry, proceed past it (see morning brief)" — a compliant autonomous agent halts, and an internal artifact ("morning brief", "Mum dogfood feedback") shipped to public users. → Rewritten to explaincheck_gateis an EXIT gate (evidence to advance, not enter), so a first-entryfalseis expected; enforced on advance. Internal refs removed./customise-workflowwrites.atlas-ai/config/atlas.json, but the engine readtoken_economyonly from.atlas-ai/fleet.json. →load_fleet_confignow honorsatlas.json's economy (fleet.json still wins if set); schema doc updated.Version hygiene
5.2.1 bumped
package.json/plugin.jsonbut missedprd_taskmaster/__init__.py(the source-of-truth the manifest tests check) — fixed here; all version locations now agree at 5.2.2.Verification
pytest tests/), CLI + MCP.tests/core/test_prerelaunch_ux_fixes.py(UX-P0-3 economy precedence).morning brief/Mum dogfood/ false-STOP strings remain in shipped skills/README/templates.Deferred (UX P1s, tracked for 5.3)
Naming maze (
atlas/prd/go/taskmaster),$SKILL_DIR/$VALIDATION_GRADEunresolved vars inhandoff, autonomous-path Claude-Code-only assumptions,execute-fleetship-check path divergence,expand-tasks/generateStep 6 overlap.🤖 Generated with Claude Code