codex usage

    Codex Weekly Usage — August 17–23, 2026

    Executive summary

    Hyphenomenon product and reporting led the observed week, with Rock Paper Scissors operations as the second major thread.

    295 active structured summaries resolve into 6 public work arcs. Review and quality assurance is the strongest observed work-mode signal. 6 outcome or decision signals were identified, with 5 follow-ups still open.

    Codex execution telemetry is supporting evidence, not a measure of productivity, duration, or the week's value.

    6
    Work arcs

    Distinct public work threads supported by retained Skysight summaries.

    6
    Outcome signals

    Public-safe outcomes or decisions supported by observed context.

    5
    Open follow-ups

    Follow-ups the observed record does not show as resolved.

    46
    Top-level tasks

    Top-level Codex tasks recorded in local execution logs; not a productivity score.

    Week in reviewObserved Aug 17, 7:30 AM–Aug 23, 6:40 PM

    What shaped the week

    Major work arcs

    Hyphenomenon product and reporting and Rock Paper Scissors operations are the two largest observed arcs by retained summary mentions. Counts can overlap and do not represent duration.

    Observed evidence by work arc

    Retained structured-summary mentions; overlapping evidence, not time allocation.

    Hyphenomenon product and reporting
    135
    Rock Paper Scissors operations
    122
    Product strategy
    57
    Research and administration
    45
    Other observed work
    25
    Mailbox and customer operations
    2

    Hyphenomenon product and reporting

    135 observed

    Product-system review, reporting strategy, and project sequencing.

    Rock Paper Scissors operations

    122 observed

    Catalog quality review and wholesale or fulfillment preparation.

    Product strategy

    57 observed

    Product framing, research, and strategic decision-making.

    3 additional work arcsLower-volume observed threads remain available without competing with the primary arcs.

    Research and administration

    45 observed

    Research, purchasing, and administrative follow-through.

    Other observed work

    25 observed

    Observed work that did not match a named public reporting arc.

    Mailbox and customer operations

    2 observed

    Mailbox, customer-support, and operational follow-through.

    Outcomes and decisions

    • Catalog image quality finding

      Advanced

      Accepted the main product image while identifying framed mockups that still needed follow-up.

    • Faire fulfillment input prepared

      Advanced

      Prepared the pick-list input for a size-counting and fulfillment task.

    • Project boundaries clarified

      Completed

      Separated consolidation, component refinement, source migration, and reporting-standard work into clearer scopes.

    • Implementation sequence authorized

      Completed

      Clarified the project sequence and authorized the next implementation steps.

    • Delivery and implementation progress

      Advanced

      Retained summaries indicate implementation, publication, deployment, or other delivery progress.

    • Review and verification decision

      Advanced

      Retained summaries indicate an approval, acceptance, verification, or passed review gate.

    Next steps and open loops

    • Framed-mockup remediation

      Open

      Determine the remaining scope and complete the framed-mockup quality remediation.

    • Pick-list size totals

      Open

      Complete and verify the size totals requested from the Faire pick list.

    • Outstanding follow-through

      Open

      Retained summaries identify work that remained pending or still needed follow-through.

    • Unresolved quality or reliability issue

      Open

      Retained summaries identify an unresolved issue, defect, failure, or remediation need.

    • Protected workflow confirmation

      Open

      One protected workflow ended without a captured final confirmation. Its subject remains excluded from the public report.

    How the observed week unfolded7 day summaries
    1. Aug 17

      Hyphenomenon product and reporting + Rock Paper Scissors operations

      53 active summaries, led by review and quality assurance signals. 15 additional observed summaries were protected.

    2. Aug 18

      Hyphenomenon product and reporting + Research and administration

      56 active summaries, led by review and quality assurance signals. 19 additional observed summaries were protected.

    3. Aug 19

      Rock Paper Scissors operations + Hyphenomenon product and reporting

      56 active summaries, led by operations and fulfillment signals. 16 additional observed summaries were protected.

    4. Aug 20

      Rock Paper Scissors operations + Other observed work

      31 active summaries, led by operations and fulfillment signals. 23 additional observed summaries were protected.

    5. Aug 21

      Hyphenomenon product and reporting + Rock Paper Scissors operations

      6 active summaries, led by review and quality assurance signals. 2 additional observed summaries were protected.

    6. Aug 22

      Hyphenomenon product and reporting + Rock Paper Scissors operations

      46 active summaries, led by review and quality assurance signals. 1 additional observed summary was protected.

    7. Aug 23

      Product strategy + Rock Paper Scissors operations

      47 active summaries, led by review and quality assurance signals. 16 additional observed summaries were protected.

    How to read this evaluationWork-mode signals, protected context, and evidence limits

    Observed work-mode signals

    Overlapping summary-level signals, not duration or a complete time sheet.

    Review and quality assurance
    145
    Operations and fulfillment
    101
    Research and administration
    58
    Building and delivery
    47
    Planning and decisions
    40

    Privacy and evidence boundary

    92 observed summaries were excluded from public topic analysis. Their time positions remain visible so the report does not mislabel protected context as missing or inactive.

    This evaluation is derived from retained Skysight titles, descriptions, application metadata, and current-window recording summaries. It describes observed evidence, not a complete time sheet, and does not infer activity inside uncovered intervals.

    Evidence appendix

    Repository activity, delivery records, source evidence, and technical diagnostics support the evaluation without competing with its first read.

    Explore supporting evidenceExecution, delivery, Skysight source context, and technical telemetry

    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.

    hyphenomenon9,641 46%
    shopify-headless5,406 26%
    multi-mailbox-ops2,102 10%
    aerial-platform1,636 8%
    Outside git metadata1,235 6%
    xyppy-chatgpt742 4%
    chronicle-visualizer20 <1%
    rps-creative-assets3 <1%

    Output Shape

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

    File Touch Mix

    File touches grouped by category; exact counts and share of retained touches.

    Docs and governance1,946 67%
    Application and pipeline code707 24%
    Other files227 8%
    Config and automation16 <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
    Count253
    Share36%
    Commit operations - success
    Count161
    Share23%
    Content push operations - success
    Count149
    Share21%
    Commit operations - unknown
    Count63
    Share9%
    Pull request operations - success
    Count28
    Share4%
    Content push operations - unknown
    Count19
    Share3%
    Branch cleanup pushes - success
    Count14
    Share2%
    Other
    Count19
    Share3%

    Repositories In Successful Push Output

    aerial-platform
    Count1
    Share20%
    chronicle-visualizer
    Count1
    Share20%
    hyphenomenon
    Count1
    Share20%
    shopify-headless
    Count1
    Share20%
    xyppy-chatgpt
    Count1
    Share20%

    Agent Stack Used

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

    App or connectorGmail
    Runtime providerNode tool runtime
    Loaded skillmulti-mailbox-ops
    • 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.

    Gmail
    Invoked7 rollouts
    256
    Figma
    Invoked3 rollouts
    228
    Vercel
    Invoked4 rollouts
    66
    Google Drive
    Invoked6 rollouts
    47
    Neon Postgres
    Invoked2 rollouts
    42
    Adobe
    Invoked1 rollout
    38
    Shopify
    Invoked4 rollouts
    18
    GitHub
    Invoked1 rollout
    5
    Resend
    Invoked1 rollout
    3
    Netlify
    Invoked1 rollout
    1

    Tool Providers

    Completed MCP provider calls, grouped by server namespace.

    Node tool runtime
    Invoked22 rollouts
    4,173
    Multi Mailbox Ops
    Invoked12 rollouts
    2,483
    Playwright browser
    Invoked11 rollouts
    679
    Computer History
    Invoked5 rollouts
    7
    OpenaiDeveloperDocs
    Invoked1 rollout
    7
    Shopify Dev Mcp
    Invoked2 rollouts
    7
    Creative Production Mcp
    Invoked1 rollout
    5

    Skills

    Explicitly requested or loaded skills; application is not inferred.

    multi-mailbox-ops
    Loaded8 rollouts
    147
    control-in-app-browser
    Loaded17 rollouts
    86
    figma-use
    Loaded5 rollouts
    58
    frontend-app-builder
    Loaded6 rollouts
    55
    control-chrome
    Loaded13 rollouts
    44
    audit
    Loaded3 rollouts
    43
    user-context
    Loaded6 rollouts
    39
    index
    Loaded6 rollouts
    38
    shopify-admin
    Loaded10 rollouts
    34
    shopify-admin-execution
    Loaded10 rollouts
    34

    Skysight Evidence Across The Week

    This evidence view locates the public work arcs and protected-context positions that support the synthesis. The calendar grid is not expected computer activity, and values are not duration.

    Open source reportSkysight Daily — Aug 20, 2026
    12–4a4–8a8a–12p12–4p4–8p8p–12a

    Hyphenomenon product and reporting

    Rock Paper Scissors operations

    Product strategy

    Research and administration

    Other observed work

    Mailbox and customer operations

    Protected context

    Color intensity represents overlapping structured-summary mentions. Empty outlined cells have no retained summary evidence; zero means retained evidence exists but this domain did not match. Values are not duration.
    Work domain12–4a4–8a8a–12p12–4p4–8p8p–12a
    Hyphenomenon product and reportingHyphenomenon product and reporting, 12–4a: 2 summary mentionsHyphenomenon product and reporting, 4–8a: 27 summary mentionsHyphenomenon product and reporting, 8a–12p: 37 summary mentionsHyphenomenon product and reporting, 12–4p: 47 summary mentionsHyphenomenon product and reporting, 4–8p: 22 summary mentionsHyphenomenon product and reporting, 8p–12a: zero matches within retained summaries
    Rock Paper Scissors operationsRock Paper Scissors operations, 12–4a: zero matches within retained summariesRock Paper Scissors operations, 4–8a: 22 summary mentionsRock Paper Scissors operations, 8a–12p: 64 summary mentionsRock Paper Scissors operations, 12–4p: 32 summary mentionsRock Paper Scissors operations, 4–8p: 4 summary mentionsRock Paper Scissors operations, 8p–12a: zero matches within retained summaries
    Product strategyProduct strategy, 12–4a: zero matches within retained summariesProduct strategy, 4–8a: 2 summary mentionsProduct strategy, 8a–12p: 8 summary mentionsProduct strategy, 12–4p: 21 summary mentionsProduct strategy, 4–8p: 26 summary mentionsProduct strategy, 8p–12a: zero matches within retained summaries
    Research and administrationResearch and administration, 12–4a: zero matches within retained summariesResearch and administration, 4–8a: 16 summary mentionsResearch and administration, 8a–12p: 10 summary mentionsResearch and administration, 12–4p: 9 summary mentionsResearch and administration, 4–8p: 10 summary mentionsResearch and administration, 8p–12a: zero matches within retained summaries
    Other observed workOther observed work, 12–4a: zero matches within retained summariesOther observed work, 4–8a: zero matches within retained summariesOther observed work, 8a–12p: 8 summary mentionsOther observed work, 12–4p: 14 summary mentionsOther observed work, 4–8p: 3 summary mentionsOther observed work, 8p–12a: zero matches within retained summaries
    Mailbox and customer operationsMailbox and customer operations, 12–4a: zero matches within retained summariesMailbox and customer operations, 4–8a: 2 summary mentionsMailbox and customer operations, 8a–12p: zero matches within retained summariesMailbox and customer operations, 12–4p: zero matches within retained summariesMailbox and customer operations, 4–8p: zero matches within retained summariesMailbox and customer operations, 8p–12a: zero matches within retained summaries
    Protected contextProtected context, 12–4a: 1 withheld summary mentionProtected context, 4–8a: 21 withheld summary mentionsProtected context, 8a–12p: 17 withheld summary mentionsProtected context, 12–4p: 26 withheld summary mentionsProtected context, 4–8p: 23 withheld summary mentionsProtected context, 8p–12a: 4 withheld summary mentions

    Structured Evidence Coverage

    Retained summaries
    402
    Active public summaries
    295
    Protected summaries
    Topic excluded; observed position retained.
    92
    Inactive or locked summaries
    15
    Covered intervals
    402
    Expected calendar intervals
    Calendar grid only; not expected computer activity.
    1,008

    Codex Execution Activity

    Token and hourly activity describe the agent execution layer. They are supporting diagnostics, not a measure of the week's value, duration, or human productivity.

    Token Usage

    Cumulative runtime counters are differenced per rollout. Repeated snapshots are ignored; cached input stays visible instead of being mistaken for fresh context.

    Token Usage

    Retained local Codex telemetry only; not an account-wide usage or billing statement.

    3,444,030,837
    Total processed
    Input + output
    Input
    3,435,372,534
    Cached input
    3,331,392,768
    Non-cached input
    103,979,766
    Output
    8,658,303
    Reasoning output
    2,573,568

    Cached input is 97% of input. Reasoning output is included within output tokens.

    Tokens By Day

    Processed input plus output assigned to the local day of each positive cumulative-counter increment.

    Wed Aug 19806,851,454 23%
    Mon Aug 17767,431,361 22%
    Tue Aug 18602,235,367 17%
    Thu Aug 20511,946,262 15%
    Sun Aug 23431,817,546 13%
    Sat Aug 22281,994,222 8%
    Fri Aug 2141,754,625 1%

    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 Aug 1710
    2. Tue Aug 187
    3. Wed Aug 1912
    4. Thu Aug 205
    5. Fri Aug 211
    6. Sat Aug 225
    7. Sun Aug 236

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

    Clock-hour pattern across the full week
    12 AM
    15
    1
    2 AM
    6
    226
    3c
    4 AM
    205
    3c
    595
    1c
    6 AM
    1,046
    14c
    1,104
    7c
    8 AM
    2,221
    18c
    1,797
    12c
    10 AM
    1,710
    12c
    1,882
    14c
    12 PM
    1,874
    13c
    1,707
    16c
    2 PM
    1,033
    9c
    1,051
    10c
    4 PM
    779
    3c
    2,024
    17c
    6 PM
    1,025
    9c
    387
    8 PM
    51
    46
    10 PM

    Total: 20,785 tool calls. “c” marks successful commit operations in that hour.

    Technical Telemetry

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

    Tool Surfaces

    Terminal/local shell20,539 99%
    Long-running task waits157 <1%
    Multi-agent tools89 <1%

    Function Calls

    wait157 64%
    wait_agent37 15%
    send_message27 11%
    followup_task9 4%
    list_agents8 3%
    spawn_agent7 3%
    interrupt_agent1 <1%

    Custom Tool Calls

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

    exec
    Count20,539
    Share99%

    Response Events

    response_item
    Count68,764
    Share53%
    event_msg
    Count58,235
    Share45%
    turn_context
    Count1,026
    Share<1%
    session_meta
    Count659
    Share<1%
    world_state
    Count376
    Share<1%
    compacted
    Count215
    Share<1%
    inter_agent_communication_metadata
    Count57
    Share<1%

    Patch Applications

    Successful patch applications
    Count1,133
    Share100%

    Output Records

    Tool output records
    Count20,752
    Share85%
    Git refs in output
    Count2,531
    Share10%
    Truncated outputs
    Count852
    Share3%
    GitHub PR links emitted
    Count269
    Share1%
    Long-running command handoffs
    Count3
    Share<1%

    Session And Log Audit

    Raw session_meta records
    Count659
    Share87%
    Active rollout log files
    Count53
    Share7%
    Unique session_meta.session_id values
    Count46
    Share6%

    Token Counter Audit

    Metered model-response increments
    24,329
    Token-count records
    24,639
    Repeated snapshots ignored
    310
    Records missing cumulative usage
    0
    Cumulative counter resets
    4

    Source Coverage

    Configured local session stores
    2
    Candidate rollout files inspected
    77
    Active rollout files
    53