UAT: `agents project context inspect` output format does not match spec — missing UKO graph visualization, active strategies, and budget breakdown #3828

New issue

Open

opened 2026-04-06 06:49:07 +00:00 by freemo · 1 comment

freemo commented

2026-04-06 06:49:07 +00:00

Owner

Metadata

Branch: fix/context-inspect-output-format-spec-alignment
Commit Message: fix(cli): align context inspect output with spec §4109-4262 — UKO graph, strategies, budget
Milestone: (backlog — no milestone assigned)
Parent Epic: #396

Background

The agents project context inspect command's output format significantly deviates from the specification's expected output.

What Was Tested

Code-level analysis of src/cleveragents/cli/commands/project_context.py context_inspect() function against docs/specification.md §4109-4262 (agents project context inspect).

Expected Behavior (from spec §4109-4262)

The spec defines the following output sections for context inspect:

UKO Graph panel: Tree visualization of the UKO subgraph rooted at focus nodes, showing hop neighborhoods with relation types (contains, imports), depth levels, and token counts per node
Active Strategies panel: Table showing each active strategy with quality score, budget percentage, and fragment count
Budget panel: Detailed breakdown including model window, response reserve, tool definitions, system prompt, skeleton reserve, available budget, and used tokens

JSON output should include:

inspection object with project, view, focus, breadth, depth, depth_name
uko_graph object with root, edges (relation, target, depth, tokens)
active_strategies array with name, quality, budget_pct, fragments
budget object with model_window, response_reserve, tool_definitions, system_prompt, skeleton_reserve, available_budget, used_tokens

Actual Behavior

The implementation (src/cleveragents/cli/commands/project_context.py, context_inspect() function) shows:

Tier Metrics table: hot_count, warm_count, cold_count, hit/miss counts
Tier Budget panel: max_tokens_hot, max_decisions_warm, max_decisions_cold
Project Fragments summary: total fragments by tier
Actor Visibility panel: fragment counts per actor role
Per-Phase Context Analysis table: fragment_count, resource_count, bytes, tokens, budget per phase
Phase Narrowing panel: narrowing checks

JSON output includes: tier_metrics, tier_budget, project_fragments, actor_visibility, acms_config, phase_analysis, fragments

Code Location

src/cleveragents/cli/commands/project_context.py — context_inspect() function (the entire rich output section)
Spec reference: docs/specification.md §4109-4262

Impact

Users cannot inspect the UKO graph structure from the CLI
Users cannot see which strategies are active and their quality/budget allocation
Users cannot see the detailed budget breakdown (model window, skeleton reserve, etc.)
The JSON output format is incompatible with any tooling that parses the spec-defined format

Subtasks

Implement UKO graph visualization in context_inspect() rich output
Add Active Strategies panel showing quality, budget_pct, fragments per strategy
Add Budget panel with model_window, response_reserve, tool_definitions, system_prompt, skeleton_reserve, available_budget, used_tokens
Update JSON output to match spec format (inspection, uko_graph, active_strategies, budget)
Add Behave tests for the new output format

Definition of Done

context inspect rich output matches spec §4109-4262 format
context inspect JSON output matches spec §4207-4262 format
All existing tests pass
New Behave tests cover the output format
All nox stages pass
Coverage >= 97%

Backlog note: This issue was discovered during autonomous UAT testing of the ACMS Context Management feature area. It does not block milestone completion and has been placed in the backlog for human review and future milestone assignment.

Automated by CleverAgents Bot
Supervisor: UAT Testing | Agent: ca-new-issue-creator

## Metadata - **Branch**: `fix/context-inspect-output-format-spec-alignment` - **Commit Message**: `fix(cli): align context inspect output with spec §4109-4262 — UKO graph, strategies, budget` - **Milestone**: *(backlog — no milestone assigned)* - **Parent Epic**: #396 ## Background The `agents project context inspect` command's output format significantly deviates from the specification's expected output. ## What Was Tested Code-level analysis of `src/cleveragents/cli/commands/project_context.py` `context_inspect()` function against `docs/specification.md` §4109-4262 (agents project context inspect). ## Expected Behavior (from spec §4109-4262) The spec defines the following output sections for `context inspect`: 1. **UKO Graph panel**: Tree visualization of the UKO subgraph rooted at focus nodes, showing hop neighborhoods with relation types (contains, imports), depth levels, and token counts per node 2. **Active Strategies panel**: Table showing each active strategy with quality score, budget percentage, and fragment count 3. **Budget panel**: Detailed breakdown including model window, response reserve, tool definitions, system prompt, skeleton reserve, available budget, and used tokens JSON output should include: - `inspection` object with project, view, focus, breadth, depth, depth_name - `uko_graph` object with root, edges (relation, target, depth, tokens) - `active_strategies` array with name, quality, budget_pct, fragments - `budget` object with model_window, response_reserve, tool_definitions, system_prompt, skeleton_reserve, available_budget, used_tokens ## Actual Behavior The implementation (`src/cleveragents/cli/commands/project_context.py`, `context_inspect()` function) shows: 1. **Tier Metrics table**: hot_count, warm_count, cold_count, hit/miss counts 2. **Tier Budget panel**: max_tokens_hot, max_decisions_warm, max_decisions_cold 3. **Project Fragments summary**: total fragments by tier 4. **Actor Visibility panel**: fragment counts per actor role 5. **Per-Phase Context Analysis table**: fragment_count, resource_count, bytes, tokens, budget per phase 6. **Phase Narrowing panel**: narrowing checks JSON output includes: `tier_metrics`, `tier_budget`, `project_fragments`, `actor_visibility`, `acms_config`, `phase_analysis`, `fragments` ## Code Location - `src/cleveragents/cli/commands/project_context.py` — `context_inspect()` function (the entire rich output section) - Spec reference: `docs/specification.md` §4109-4262 ## Impact - Users cannot inspect the UKO graph structure from the CLI - Users cannot see which strategies are active and their quality/budget allocation - Users cannot see the detailed budget breakdown (model window, skeleton reserve, etc.) - The JSON output format is incompatible with any tooling that parses the spec-defined format ## Subtasks - [ ] Implement UKO graph visualization in `context_inspect()` rich output - [ ] Add Active Strategies panel showing quality, budget_pct, fragments per strategy - [ ] Add Budget panel with model_window, response_reserve, tool_definitions, system_prompt, skeleton_reserve, available_budget, used_tokens - [ ] Update JSON output to match spec format (inspection, uko_graph, active_strategies, budget) - [ ] Add Behave tests for the new output format ## Definition of Done - [ ] `context inspect` rich output matches spec §4109-4262 format - [ ] `context inspect` JSON output matches spec §4207-4262 format - [ ] All existing tests pass - [ ] New Behave tests cover the output format - All nox stages pass - Coverage >= 97% > **Backlog note:** This issue was discovered during autonomous UAT testing of the ACMS Context Management feature area. It does not block milestone completion and has been placed in the backlog for human review and future milestone assignment. --- **Automated by CleverAgents Bot** Supervisor: UAT Testing | Agent: ca-new-issue-creator

freemo added a new dependency

2026-04-06 06:49:33 +00:00

#396 Epic: ACMS Context Pipeline

freemo commented

2026-04-06 06:51:07 +00:00

Author

Owner

Label compliance fix applied:

Removed conflicting label: State/Unverified
Reason: Issue had both State/In Progress and State/Unverified. Keeping State/In Progress as the more advanced state.

Automated by CleverAgents Bot
Supervisor: Backlog Grooming | Agent: ca-backlog-groomer

Label compliance fix applied: - Removed conflicting label: `State/Unverified` - Reason: Issue had both `State/In Progress` and `State/Unverified`. Keeping `State/In Progress` as the more advanced state. --- **Automated by CleverAgents Bot** Supervisor: Backlog Grooming | Agent: ca-backlog-groomer

freemo referenced this issue

2026-04-06 06:51:48 +00:00

[Automated] CleverAgents Build Session - 2026-04-06 #3775

freemo referenced this issue

2026-04-06 06:51:56 +00:00

[Automated] CleverAgents Build Session - 2026-04-06 #3775

freemo referenced this issue

2026-04-06 07:04:26 +00:00

[Automated] CleverAgents Build Session - 2026-04-06 #3775

freemo referenced this issue

2026-04-06 10:35:53 +00:00

[Automated] CleverAgents Build Session - 2026-04-06 #3775