UAT: Built-in automation profile descriptions for `manual`, `review`, `supervised`, `cautious`, `trusted`, and `auto` don't match spec #4695

New issue

Open

opened 2026-04-08 18:02:01 +00:00 by HAL9000 · 0 comments

HAL9000 commented

2026-04-08 18:02:01 +00:00

Owner

Bug Report

Feature Area: Automation Profiles — Built-in Profile Definitions
Severity: Low — description text mismatch (extends #4594 which only covered ci and full-auto)
Discovered by: UAT worker uat-tester-automation-profiles

What Was Tested

The description strings of all 8 built-in automation profiles against the spec-defined descriptions in the automation-profile list output (spec lines 17040–17047) and JSON output (spec lines 17095–17102).

Expected Behavior (from spec)

Per the specification (§ agents automation-profile list, lines 17040–17047 and JSON output at lines 17095–17102), the descriptions for all built-in profiles are:

Profile	Expected Description
`manual`	`"Human-driven (default)"`
`review`	`"Auto-phase, manual decisions"`
`supervised`	`"Auto-plan, manual exec"`
`cautious`	`"Confidence-gated automation"`
`trusted`	`"Auto-exec, manual apply"`
`auto`	`"Full auto except apply"`
`ci`	`"Full auto, safe CI/CD"` ← already filed in #4594
`full-auto`	`"Complete automation"` ← already filed in #4594

Note: The trusted profile's description in the show output (spec line 17274) is "Auto-exec, manual apply. Day-to-day development" — the list description is the shorter form.

Actual Behavior

The implementation in src/cleveragents/domain/models/core/automation_profile.py (lines 426–609) has:

Profile	Actual Description
`manual`	`"Human approves every action"`
`review`	`"Human reviews before apply"`
`supervised`	`"Human reviews strategy and execution"`
`cautious`	`"Probabilistic gates on most actions"`
`trusted`	`"Auto for most, human for apply and revert"`
`auto`	`"Fully automatic except apply"`

(Note: ci and full-auto are already tracked in issue #4594.)

Code Location

src/cleveragents/domain/models/core/automation_profile.py, lines 426–609 (BUILTIN_PROFILES dict)

Fix Required

Update the description fields in BUILTIN_PROFILES:

"manual": AutomationProfile(
    name="manual",
    description="Human-driven (default)",  # was: "Human approves every action"
    ...
),
"review": AutomationProfile(
    name="review",
    description="Auto-phase, manual decisions",  # was: "Human reviews before apply"
    ...
),
"supervised": AutomationProfile(
    name="supervised",
    description="Auto-plan, manual exec",  # was: "Human reviews strategy and execution"
    ...
),
"cautious": AutomationProfile(
    name="cautious",
    description="Confidence-gated automation",  # was: "Probabilistic gates on most actions"
    ...
),
"trusted": AutomationProfile(
    name="trusted",
    description="Auto-exec, manual apply",  # was: "Auto for most, human for apply and revert"
    ...
),
"auto": AutomationProfile(
    name="auto",
    description="Full auto except apply",  # was: "Fully automatic except apply"
    ...
),

Impact

The automation-profile list output will show wrong descriptions for 6 of the 8 built-in profiles
The JSON output from automation-profile list --format json will have wrong description values
Any tooling that parses the description field will get incorrect values
The automation-profile show output will show wrong descriptions

Automated by CleverAgents Bot
Supervisor: UAT Testing | Agent: uat-tester

## Bug Report **Feature Area:** Automation Profiles — Built-in Profile Definitions **Severity:** Low — description text mismatch (extends #4594 which only covered `ci` and `full-auto`) **Discovered by:** UAT worker uat-tester-automation-profiles --- ## What Was Tested The description strings of all 8 built-in automation profiles against the spec-defined descriptions in the `automation-profile list` output (spec lines 17040–17047) and JSON output (spec lines 17095–17102). ## Expected Behavior (from spec) Per the specification (§ `agents automation-profile list`, lines 17040–17047 and JSON output at lines 17095–17102), the descriptions for all built-in profiles are: | Profile | Expected Description | |---------|---------------------| | `manual` | `"Human-driven (default)"` | | `review` | `"Auto-phase, manual decisions"` | | `supervised` | `"Auto-plan, manual exec"` | | `cautious` | `"Confidence-gated automation"` | | `trusted` | `"Auto-exec, manual apply"` | | `auto` | `"Full auto except apply"` | | `ci` | `"Full auto, safe CI/CD"` ← already filed in #4594 | | `full-auto` | `"Complete automation"` ← already filed in #4594 | Note: The `trusted` profile's description in the `show` output (spec line 17274) is `"Auto-exec, manual apply. Day-to-day development"` — the list description is the shorter form. ## Actual Behavior The implementation in `src/cleveragents/domain/models/core/automation_profile.py` (lines 426–609) has: | Profile | Actual Description | |---------|-------------------| | `manual` | `"Human approves every action"` | | `review` | `"Human reviews before apply"` | | `supervised` | `"Human reviews strategy and execution"` | | `cautious` | `"Probabilistic gates on most actions"` | | `trusted` | `"Auto for most, human for apply and revert"` | | `auto` | `"Fully automatic except apply"` | (Note: `ci` and `full-auto` are already tracked in issue #4594.) ## Code Location `src/cleveragents/domain/models/core/automation_profile.py`, lines 426–609 (`BUILTIN_PROFILES` dict) ## Fix Required Update the `description` fields in `BUILTIN_PROFILES`: ```python "manual": AutomationProfile( name="manual", description="Human-driven (default)", # was: "Human approves every action" ... ), "review": AutomationProfile( name="review", description="Auto-phase, manual decisions", # was: "Human reviews before apply" ... ), "supervised": AutomationProfile( name="supervised", description="Auto-plan, manual exec", # was: "Human reviews strategy and execution" ... ), "cautious": AutomationProfile( name="cautious", description="Confidence-gated automation", # was: "Probabilistic gates on most actions" ... ), "trusted": AutomationProfile( name="trusted", description="Auto-exec, manual apply", # was: "Auto for most, human for apply and revert" ... ), "auto": AutomationProfile( name="auto", description="Full auto except apply", # was: "Fully automatic except apply" ... ), ``` ## Impact - The `automation-profile list` output will show wrong descriptions for 6 of the 8 built-in profiles - The JSON output from `automation-profile list --format json` will have wrong `description` values - Any tooling that parses the description field will get incorrect values - The `automation-profile show` output will show wrong descriptions --- **Automated by CleverAgents Bot** Supervisor: UAT Testing | Agent: uat-tester

HAL9000 added this to the v3.5.0 milestone

2026-04-08 18:05:36 +00:00