codex usage
    Template-generated · codex-usage-report-v5Model · Not recorded

    Codex Weekly Usage — July 20–26, 2026

    A storefront-heavy week spanning Shopify Headless and Hyphenomenon: 17 de-duplicated Codex sessions across all seven days, balanced application and documentation work, verified delivery traces for two repositories, and a complete Chronicle coverage gap.

    17
    Unique Codex sessions

    De-duplicated by session_meta.session_id; 17 active rollout log files are tracked separately.

    90
    Successful commit operations

    Verified from terminal outcomes; operations are not independent feature units.

    311
    Files touched

    Unique paths in successful patch-application records.

    2
    Repositories delivered

    hyphenomenon and shopify-headless

    Activity By Day And Clock Hour

    The day row shows each date in the report window. The hour chart below combines the same clock hour across all seven days; it is not a single-day timeline.

    Daily session timeline
    1. Mon Jul 207
    2. Tue Jul 211
    3. Wed Jul 222
    4. Thu Jul 231
    5. Fri Jul 241
    6. Sat Jul 252
    7. Sun Jul 263

    Each count is a de-duplicated session start assigned to that local calendar day.

    Clock-hour pattern across the full week
    1
    12 AM
    2 AM
    313
    5c
    4 AM
    301
    4c
    182
    1c
    6 AM
    791
    10c
    522
    8c
    8 AM
    584
    4c
    1,291
    21c
    10 AM
    481
    6c
    668
    8c
    12 PM
    1,145
    11c
    314
    5c
    2 PM
    315
    2c
    151
    2c
    4 PM
    6
    134
    2c
    6 PM
    175
    1c
    5
    8 PM
    4
    1
    10 PM

    Total: 7,384 tool calls. “c” marks successful commit operations in that hour.

    Where The Work Landed

    Repository activity and changed lines show which codebases received attention during the report window.

    Tool Calls By Repository

    Working repository resolved from tool workdir or turn context.

    shopify-headless4,120 56%
    hyphenomenon2,866 39%
    Outside git metadata385 5%
    codex-skills12 <1%
    chronicle-visualizer1 <1%

    Patch Churn By Repository

    Added plus removed lines recorded in patch inputs.

    shopify-headless21,971 64%
    hyphenomenon12,107 35%
    Outside git metadata278 <1%

    Output Shape

    Compare the rollouts carrying the largest patches with the mix of file categories touched.

    Patch Churn By Rollout

    Rollout 1015,514 45%

    Top-level | shopify-headless

    Rollout 156,606 19%

    Top-level | hyphenomenon

    Rollout 136,214 18%

    Top-level | shopify-headless

    Rollout 145,329 16%

    Top-level | hyphenomenon

    Rollout 17278 <1%

    Top-level | Outside git metadata

    Rollout 11243 <1%

    Top-level | shopify-headless

    Rollout 01172 <1%

    Top-level | hyphenomenon

    File Touch Mix

    1,413touches
    Docs and governance70050%
    Application and pipeline code68248%
    Other files282%
    Config and automation3<1%

    Verified Delivery Ledger

    Outcomes come from terminal return records. Intent-only command counts remain in the technical appendix.

    Command Outcomes

    Pull request operations - unknown
    Count162
    Share38%
    Commit operations - success
    Count90
    Share21%
    Content push operations - success
    Count80
    Share19%
    Pull request operations - success
    Count35
    Share8%
    Commit operations - unknown
    Count22
    Share5%
    Content push operations - unknown
    Count15
    Share3%
    Pull request operations - pending
    Count6
    Share1%
    Other
    Count20
    Share5%

    Repositories In Successful Push Output

    hyphenomenon
    Count1
    Share50%
    shopify-headless
    Count1
    Share50%

    Verification Was Part Of The Work

    Terminal records include 90 successful commit operations, 80 successful content-push operations, 35 successful pull-request operations, and three successful branch-cleanup pushes. The same evidence retains unknown, pending, and failed outcomes—including five content pushes that ended in long-running handoffs—so it would be wrong to claim every attempted operation completed. Operation counts are execution evidence, not independent feature units.

    Agent Stack Used

    Observed runtime evidence only; the available catalog is excluded. Status chips distinguish completed calls, skill loads, and explicit requests.

    App or connectorShopify
    Runtime providerNode tool runtime
    Loaded skillcontrol-in-app-browser
    • Invoked means a completed runtime call was recorded.
    • Loaded means a skill file was read or discovered.
    • Requested marks an explicit skill request in the source.
    • The logs do not preserve a reliable causal chain between every layer, so the diagram shows observed layers rather than inventing one.

    Apps and Connectors

    Completed connector calls recorded by the runtime.

    Shopify
    Invoked2 rollouts
    290
    Vercel
    Invoked4 rollouts
    189
    Gmail
    Invoked3 rollouts
    51
    Figma
    Invoked1 rollout
    45
    GitHub
    Invoked3 rollouts
    7

    Tool Providers

    Completed MCP provider calls, grouped by server namespace.

    Node tool runtime
    Invoked10 rollouts
    1,164
    Playwright browser
    Invoked3 rollouts
    134
    Shopify Dev Mcp
    Invoked2 rollouts
    98
    Creative Production Mcp
    Invoked1 rollout
    24

    Skills

    Explicitly requested or loaded skills; application is not inferred.

    control-in-app-browser
    Loaded6 rollouts
    42
    frontend-app-builder
    Loaded6 rollouts
    34
    frontend-testing-debugging
    Loaded4 rollouts
    20
    project-lifecycle
    Loaded3 rollouts
    14
    github
    Loaded3 rollouts
    13
    motion-design-director
    Loaded2 rollouts
    13
    react-best-practices
    Loaded4 rollouts
    13
    hyphenomenon-codex-usage-report
    Loaded + Requested2 rollouts
    9
    control-chrome
    Loaded2 rollouts
    7
    continue-until-blocked
    Loaded + Requested3 rollouts
    6

    Technical Telemetry

    Raw tool, event, workspace, and patch diagnostics used to audit the analyzer. These counts are intentionally secondary to outcomes.

    Analyzer artifactCodex Usage Report Analyzer v5 Version lineageHow This Site Works

    Tool Surfaces

    Terminal/local shell5,067 69%
    Node/browser automation929 13%
    Code editing453 6%
    Long-running task waits276 4%
    Vercel176 2%
    Other/uncategorized162 2%
    Local image review146 2%
    Other175 2%

    Function Calls

    1,050
    929
    620
    276
    146
    69
    55
    389
    exec_command
    js
    write_stdin
    wait
    view_image
    _get_deployment
    update_plan
    Other

    Custom Tool Calls

    Specific custom calls and their share of all recorded tool events.

    exec
    Count3,397
    Share46%
    apply_patch
    Count453
    Share6%

    Response Events

    response_item
    Count23,663
    Share61%
    event_msg
    Count14,536
    Share38%
    turn_context
    Count222
    Share<1%
    session_meta
    Count158
    Share<1%
    world_state
    Count124
    Share<1%
    compacted
    Count56
    Share<1%

    Patch Applications

    Successful patch applications
    Count603
    Share57%
    Apply patch calls
    Count453
    Share43%

    Output Records

    Tool output records
    Count7,382
    Share79%
    Git refs in output
    Count1,009
    Share11%
    Long-running command handoffs
    Count428
    Share5%
    Truncated outputs
    Count230
    Share2%
    GitHub PR links emitted
    Count154
    Share2%
    Non-zero terminal exits
    Count102
    Share1%

    Session And Log Audit

    Raw session_meta records
    Count158
    Share82%
    Active rollout log files
    Count17
    Share9%
    Unique session_meta.session_id values
    Count17
    Share9%