Skip to content

thedotmack/claude-mem / plugin/skills/agent-cost-report

agent-cost-report

Verified

Believable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.

FileKindSize
.gitignoreother100 B
CHECKSUMS.txtmarkdown2.1 KB
SKILL.mdskill md21.3 KB
VERIFICATION.mdmarkdown12.3 KB
scripts/acr.pyexecutablescript12.2 KB
scripts/acr/__init__.pyscript105 B
scripts/acr/behavior.pyscript17.5 KB
scripts/acr/classify.pyscript5.5 KB
scripts/acr/costs.pyscript6.3 KB
scripts/acr/devices.pyscript5.4 KB
scripts/acr/evidence.pyscript9.2 KB
scripts/acr/labels.pyscript7.6 KB
scripts/acr/measure.pyscript5.6 KB
scripts/acr/mistakes.pyscript23.1 KB
scripts/acr/owner.pyscript2.9 KB
scripts/acr/patterns.pyscript26.0 KB
scripts/acr/period.pyscript3.8 KB
scripts/acr/prices.pyscript5.0 KB
scripts/acr/render.pyscript52.7 KB
scripts/acr/rollup.pyscript31.7 KB
scripts/acr/rules.jsonconfig3.5 KB
scripts/acr/rules.pyscript4.5 KB
scripts/acr/sample.pyscript3.6 KB
scripts/acr/snapshot.pyscript2.0 KB
scripts/acr/sync.pyscript3.7 KB
scripts/acr/transcripts.pyscript7.4 KB
scripts/acr/wins.pyscript21.0 KB
scripts/tests/_db.pyscript3.6 KB
scripts/tests/_paths.pyscript337 B
scripts/tests/fixtures/codex/sessions/2026/09/18/rollout-codex-1.jsonlother664 B
scripts/tests/fixtures/make_report_fixtures.pyscript2.0 KB
scripts/tests/fixtures/prices_sample.jsonconfig611 B
scripts/tests/fixtures/projects/-workspace-proj-a/sess-aaaa.jsonlother2.9 KB
scripts/tests/fixtures/projects/-workspace-proj-b/sess-bbbb.jsonlother537 B
scripts/tests/fixtures/report-1day.jsonconfig20.1 KB
scripts/tests/fixtures/report-30day.jsonconfig48.1 KB
scripts/tests/fixtures/report-7day.jsonconfig29.1 KB
scripts/tests/fixtures/report-empty.jsonconfig13.8 KB
scripts/tests/fixtures/report-session.jsonconfig16.6 KB
scripts/tests/test_behavior.pyscript29.6 KB
scripts/tests/test_devices.pyscript5.9 KB
scripts/tests/test_labels.pyscript8.6 KB
scripts/tests/test_measure.pyscript6.8 KB
scripts/tests/test_period.pyscript7.0 KB
scripts/tests/test_prices.pyscript5.3 KB
scripts/tests/test_render.pyscript12.1 KB
scripts/tests/test_render_encoding.pyscript1.5 KB
scripts/tests/test_rollup.pyscript19.3 KB
scripts/tests/test_snapshot.pyscript2.8 KB
scripts/tests/test_sync.pyscript2.5 KB
scripts/tests/test_transcript_encoding.pyscript2.7 KB
scripts/tests/test_transcripts.pyscript8.1 KB
scripts/tests/test_wins.pyscript15.7 KB
verification/precision-2026-09-26.jsonconfig308 B

scripts/tests/fixtures/report-1day.json

View on GitHub
{
 "attention": [],
 "behavior": {
  "author_tags": {
   "bot": 0,
   "human": 1,
   "unknown": 0
  },
  "classifier": {
   "cap_usd": 2.0,
   "model": null,
   "note": "off by default (G8); enable with --classify",
   "ran": false,
   "sampled_n": 0,
   "spend_usd": null
  },
  "coverage": {
   "box": "measured",
   "grok_bot": "unavailable",
   "mac": "unavailable (no Phase 5 export)"
  },
  "default_model": "anthropic/claude-sonnet-5",
  "episodes": [],
  "episodes_status": "measured",
  "high_turns_n": 0,
  "human_message_counts": "computed from human-tagged turns",
  "label_sources": {},
  "mistake_clusters": [],
  "not_in_mistakes_line": "session-level keyword failure signals with no turn span stay in Details (2B.9)",
  "outbound": {
   "alex_minutes": null,
   "incidents": 0,
   "recipients": 0
  },
  "patterns": [
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "medium",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "Unnecessary escalation",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P1_invented_gates",
    "low_usd": 0.0,
    "name": "Invented human gates / asking instead of doing",
    "never_answered_n": 0,
    "overlaps": true,
    "placement": "tile",
    "source_ref": "metrics-ideas.md:16",
    "unconfirmed_n": 0,
    "unmeasured_n": 0,
    "waiting_minutes": 0
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "medium",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "Regression",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P2_broke_things",
    "low_usd": 0.0,
    "name": "Broke working things",
    "overlaps": true,
    "placement": "tile",
    "source_ref": "metrics-ideas.md:17",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "high",
    "count": 0,
    "count_basis": "heuristic",
    "default_model": "anthropic/claude-sonnet-5",
    "delta_usd": 0.0,
    "detection": "H",
    "examples": [],
    "failure_type": "Model thrash",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P3_wrong_model",
    "low_usd": 0.0,
    "model_not_logged": 0,
    "name": "Wrong or expensive model",
    "overlaps": true,
    "placement": "tile",
    "source_ref": "metrics-ideas.md:18",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "medium",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "Suboptimal path",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P4_over_engineering",
    "low_usd": 0.0,
    "name": "Over-engineering / overthinking",
    "overlaps": true,
    "placement": "details",
    "source_ref": "metrics-ideas.md:19",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "medium",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "Missed requirement",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P5_not_asked",
    "low_usd": 0.0,
    "name": "Did something not asked",
    "overlaps": true,
    "placement": "details",
    "source_ref": "metrics-ideas.md:20",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "medium",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "Premature completion",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P6_fake_output",
    "low_usd": 0.0,
    "name": "Fake or invented output",
    "overlaps": true,
    "placement": "tile",
    "source_ref": "metrics-ideas.md:21",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "medium",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "Premature completion",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P7_false_done",
    "low_usd": 0.0,
    "name": "Skipped steps / false done",
    "overlaps": true,
    "placement": "tile",
    "source_ref": "metrics-ideas.md:22",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "medium",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "Suboptimal path",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P8_wrong_tool",
    "low_usd": 0.0,
    "name": "Wrong tool, account, or contact",
    "overlaps": true,
    "placement": "details",
    "source_ref": "metrics-ideas.md:23",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "high",
    "cost_note": "incidents \u00d7 recipients, never dollars",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "Unauthorized action",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": null,
    "incidents": 0,
    "key": "P9_bad_outbound",
    "low_usd": null,
    "name": "Bad outbound",
    "overlaps": true,
    "placement": "details",
    "recipients": 0,
    "source_ref": "metrics-ideas.md:24",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "medium",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P10_memory_loss",
    "low_usd": 0.0,
    "name": "Memory or rule loss",
    "overlaps": true,
    "placement": "details",
    "source_ref": "metrics-ideas.md:25",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "medium",
    "cost_note": "Alex-minutes = words \u00f7 200; dollars only for a clarification round trip",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P11_jargon",
    "low_usd": 0.0,
    "name": "Jargon / walls of text",
    "overlaps": true,
    "placement": "details",
    "source_ref": "metrics-ideas.md:26",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "medium",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "base detector; M attributes when it can",
    "examples": [],
    "failure_type": "",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "P12_unclear",
    "low_usd": 0.0,
    "name": "Unclear cause",
    "overlaps": true,
    "placement": "details",
    "source_ref": "metrics-ideas.md:27",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "confidence": "high",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "error_chains": 0,
    "examples": [],
    "failure_type": "Recovery after miss",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "S1_tool_errors",
    "loops": 0,
    "low_usd": 0.0,
    "name": "Tool errors and retries",
    "overlaps": true,
    "placement": "details",
    "recovered_usd": 0.0,
    "source_ref": "plan 2B.3 supporting metric",
    "unmeasured_n": 0,
    "waiting_minutes": null
   },
   {
    "agent_minutes": null,
    "alex_minutes": 0,
    "availability_unknown_n": 0,
    "confidence": "medium",
    "cost_note": "text only, not priced",
    "count": 0,
    "count_basis": "heuristic",
    "detection": "H",
    "examples": [],
    "failure_type": "Hedging",
    "high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
    "high_usd": 0.0,
    "key": "S2_hedging",
    "low_usd": 0.0,
    "name": "Hedging",
    "overlaps": true,
    "placement": "details",
    "source_ref": "plan 2B.3 supporting metric",
    "unmeasured_n": 0,
    "waiting_minutes": null
   }
  ],
  "permission_denials": {
   "count": 0,
   "examples": [],
   "note": "blocked by a rule; not priced as agent waste"
  },
  "prompt_rows_n": 1,
  "rule_effectiveness": [
   {
    "after": {
     "episodes": 0,
     "human_prompts": 1,
     "per_100": null
    },
    "before": {
     "episodes": 0,
     "human_prompts": 0,
     "per_100": null
    },
    "card": false,
    "episode_ids_after": [],
    "landed_pt": "2026-09-12",
    "n_days": 7,
    "name": "All house automations OFF (credit burn)",
    "pattern": "P3_wrong_model",
    "rule_key": "automations_off",
    "source": "rules.json",
    "verdict": "not_enough_data"
   },
   {
    "after": {
     "episodes": 0,
     "human_prompts": 1,
     "per_100": null
    },
    "before": {
     "episodes": 0,
     "human_prompts": 0,
     "per_100": null
    },
    "card": false,
    "episode_ids_after": [],
    "landed_pt": "2026-09-14",
    "n_days": 7,
    "name": "Investor prep = person intel only, no lectures",
    "pattern": "P11_jargon",
    "rule_key": "investor_prep_person_intel",
    "source": "rules.json",
    "verdict": "not_enough_data"
   },
   {
    "after": {
     "episodes": 0,
     "human_prompts": 1,
     "per_100": null
    },
    "before": {
     "episodes": 0,
     "human_prompts": 0,
     "per_100": null
    },
    "card": false,
    "episode_ids_after": [],
    "landed_pt": "2026-09-15",
    "n_days": 7,
    "name": "Seats escalate to Prioritizer only, never ping Alex",
    "pattern": "P1_invented_gates",
    "rule_key": "seats_escalate_prioritizer",
    "source": "rules.json",
    "verdict": "not_enough_data"
   }
  ],
  "scan": {
   "files_seen": 0
  },
  "session_tags": {},
  "tagger_ran": true,
  "unconfirmed": {
   "P1_candidates": []
  },
  "union_high_usd": 0.0,
  "union_low_usd": 0.0,
  "union_turns_n": 0
 },
 "by_category": [
  {
   "active_minutes": 42.0,
   "agent_estimated_usd": 0.09,
   "category": "Feature",
   "extrapolated_usd": 0.0,
   "finished": 1,
   "recovery_usd": 0.0,
   "sessions": 1,
   "wasted_usd": 0.0
  }
 ],
 "by_day": [
  {
   "agent_calls": 2,
   "agent_estimated_usd": 0.09,
   "day_pt": "2026-09-18",
   "events": 13,
   "extrapolated_usd": 0.0,
   "finished_outcomes": 1,
   "no_agent_work": false,
   "sessions_started": 2,
   "ship_events": 1,
   "tokens": 55200
  }
 ],
 "by_device": [
  {
   "agent_estimated_usd": 0.09,
   "device": "local",
   "extrapolated_usd": 0.0,
   "id_hash": "local",
   "sessions": 1
  }
 ],
 "by_model": [
  {
   "agent_estimated_usd": 1.59,
   "cache_read": 50000,
   "cache_write_1h": 1000,
   "cache_write_5m": 2000,
   "calls": 4,
   "input": 102100,
   "model": "claude-fable-5-1",
   "output": 10210,
   "priced": true,
   "prices_usd_per_mtok": {
    "cache_read": 0.25,
    "cache_write": 12.5,
    "cache_write_1h": 20.0,
    "input": 10.0,
    "key": "anthropic/claude-fable-5.1",
    "output": 50.0,
    "price_source": {
     "cache_read": "listed",
     "cache_write": "listed",
     "cache_write_1h": "input*2"
    }
   },
   "source": "claude-code",
   "unpriced_calls": 0
  }
 ],
 "devices": [
  {
   "device": "local",
   "id_hash": "local"
  }
 ],
 "failure_economics": [],
 "generated_at_pt": "2026-09-25 12:00 PT",
 "labels": {
  "reviewed": 0,
  "total": 2
 },
 "line_items": [
  {
   "active_minutes": 42.0,
   "agent_calls": 2,
   "agent_tokens": {
    "cache_read": 50000,
    "cache_write_1h": 1000,
    "cache_write_5m": 2000,
    "input": 2000,
    "output": 200
   },
   "attributed_usd": 0.09,
   "behavior_counts": {},
   "category": "Feature",
   "confidence": "high",
   "content_session_id": "cs1",
   "cost_basis": "estimated_usage",
   "cost_estimated": 0.09,
   "cost_extrapolated": null,
   "cost_measured": "unavailable",
   "cost_x1e6": 87500,
   "date_pt": "2026-09-18",
   "device": [
    "local"
   ],
   "evidence_ids": [
    1,
    2,
    3,
    4
   ],
   "extrapolated_x1e6": 0,
   "failure_signals": [],
   "failure_type": "",
   "has_transcript": true,
   "label_source": "keyword",
   "model": "claude-fable-5-1",
   "model_prices_usd_per_mtok": {
    "cache_read": 0.25,
    "cache_write": 12.5,
    "cache_write_1h": 20.0,
    "input": 10.0,
    "key": "anthropic/claude-fable-5.1",
    "output": 50.0,
    "price_source": {
     "cache_read": "listed",
     "cache_write": "listed",
     "cache_write_1h": "input*2"
    }
   },
   "notes": "shipped; 4 observations, 1 summaries, 1 prompts; tokens measured from the transcript on this box",
   "observations": 4,
   "observer_tokens": 12000,
   "platform": "claude",
   "productive_cost": 0.09,
   "project": "claude-mem",
   "recommended_action": "",
   "recovery_cost": 0.0,
   "risk_exposure": "none",
   "scope": "period",
   "session_ids": [
    "m1"
   ],
   "session_status": "completed",
   "ship_events": 1,
   "ship_titles": [
    "PR #4125 merged to main"
   ],
   "status": "shipped",
   "summaries": 1,
   "summary_ids": [
    1
   ],
   "title": "implement the rollup",
   "trivial": false,
   "unpriced_calls": 0,
   "wasted_cost": 0.0,
   "work_category": "Feature",
   "work_item_id": "WI-1",
   "worktree": ""
  },
  {
   "active_minutes": 0,
   "agent_calls": 0,
   "agent_tokens": {
    "cache_read": 0,
    "cache_write_1h": 0,
    "cache_write_5m": 0,
    "input": 0,
    "output": 0
   },
   "attributed_usd": 0.0,
   "behavior_counts": {},
   "category": "Investigation",
   "confidence": "low",
   "content_session_id": "cs3",
   "cost_basis": "extrapolated",
   "cost_estimated": null,
   "cost_extrapolated": 0.0,
   "cost_measured": "unavailable",
   "cost_x1e6": 0,
   "date_pt": "2026-09-18",
   "device": [],
   "evidence_ids": [],
   "extrapolated_x1e6": 0,
   "failure_signals": [],
   "failure_type": "",
   "has_transcript": false,
   "label_source": "keyword",
   "model": null,
   "model_prices_usd_per_mtok": null,
   "notes": "abandoned; 0 observations, 0 summaries, 0 prompts; no transcript on this box (remote device)",
   "observations": 0,
   "observer_tokens": 0,
   "platform": "claude",
   "productive_cost": 0.0,
   "project": "helper",
   "recommended_action": "",
   "recovery_cost": 0.0,
   "risk_exposure": "none",
   "scope": "period",
   "session_ids": [
    "m3"
   ],
   "session_status": "completed",
   "ship_events": 0,
   "ship_titles": [],
   "status": "abandoned",
   "summaries": 0,
   "summary_ids": [],
   "title": "",
   "trivial": true,
   "unpriced_calls": 0,
   "wasted_cost": 0.0,
   "work_category": "Investigation",
   "work_item_id": "WI-2",
   "worktree": ""
  }
 ],
 "observer": {
  "basis": "ESTIMATED: observer model's own tokens (deduped per reply) x its input list price; kept separate from the agent figures above",
  "est_usd": 0.0,
  "models": [
   "gemini-2.5-flash-lite"
  ],
  "rows_in_scope": {
   "observations": 4,
   "prompts": 1,
   "summaries": 1,
   "tool_uses": 1
  },
  "tokens_dedup": 12000,
  "unpriced_tokens": {}
 },
 "pricing": {
  "fetched": "2026-09-25T17:00:00-07:00",
  "loaded_from": null,
  "models_listed": 7,
  "rule_1h": "explicit input_cache_write_1h when listed, else 2 x input (weekly_report.py:95)",
  "source": "https://openrouter.ai/api/v1/models"
 },
 "scope": {
  "kind": "period",
  "project": null,
  "session": null
 },
 "spend": {
  "agent_estimated_label": "ESTIMATED",
  "agent_estimated_usd": 1.59,
  "agent_measured_usd": null,
  "extrapolated_label": "EXTRAPOLATED (low confidence)",
  "extrapolated_unmeasured_usd": 0.0,
  "extrapolation_basis": "EXTRAPOLATED (low confidence): on this box, measured sessions cost $7.29 API-equivalent per 1M observer tokens; applied to 0 observer tokens from 1 sessions with no transcript here (remote devices). Assumes remote work looks like box work.",
  "grok_bot_usage": {
   "status": "unavailable"
  },
  "headline_label": "estimated at OpenRouter list prices from measured tokens",
  "headline_usd": 1.59,
  "measured_status": "unavailable",
  "mistakes_alex_minutes": 0,
  "mistakes_alex_minutes_known_n": 0,
  "mistakes_basis": "same-session wasted turns, each turn once",
  "mistakes_episodes_n": 0,
  "mistakes_estimated_usd": 0.0,
  "mistakes_high_label": "upper bound, likely 5\u201310\u00d7 too high (Frustration Arc)",
  "mistakes_high_usd": 0.0,
  "mistakes_label": "ESTIMATED",
  "mistakes_turns_n": 0,
  "mistakes_unmatched_usd": 0.0,
  "mistakes_unmeasured_n": 0,
  "observer_note_taker_est_usd": 0.0,
  "total_estimate_usd": 1.59
 },
 "timeline": {
  "mistakes_by_day": [
   {
    "alex_minutes": 0.0,
    "by_pattern": {},
    "day_pt": "2026-09-18",
    "high_usd": 0.0,
    "outbound_incidents": 0,
    "top_examples": [],
    "turns_n": 0,
    "unmeasured_n": 0,
    "usd": 0.0
   }
  ],
  "wins_by_day": [
   {
    "count": 1,
    "day_pt": "2026-09-18",
    "usd": null,
    "win_ids": [
     "W-1"
    ]
   }
  ]
 },
 "totals": {
  "agent_hours": 0.7,
  "cost_per_completed_outcome": 0.09,
  "devices": 1,
  "finished_outcomes": 1,
  "hours_method": "sum of gaps <=15 min between consecutive events (prompts, observations, summaries, tool calls, transcript API calls); parallel sessions add up in agent_hours but not in wall_clock_hours",
  "projects": 2,
  "real_work_sessions": 1,
  "recovery_share": 0.0,
  "recovery_usd": 0.0,
  "repos": 2,
  "sessions": 2,
  "ship_events": 1,
  "tokens": {
   "api_calls": 4,
   "cache_read": 50000,
   "cache_write_1h": 1000,
   "cache_write_5m": 2000,
   "input": 102100,
   "output": 10210,
   "total": 165310
  },
  "trivial_sessions": 1,
  "wall_clock_hours": 0.7,
  "waste_rate": 0.0,
  "wasted_usd": 0.0
 },
 "trivial_definition": "no observations, no transcript on this box, and <1 active minute",
 "trivial_sessions": [
  {
   "session_ids": [
    "m3"
   ],
   "title": "",
   "work_item_id": "WI-2"
  }
 ],
 "unmatched_transcripts": {
  "by_dir": {
   "-w-p": 2
  },
  "count": 2,
  "note": "transcripts on this box that matched no sdk_sessions row (work claude-mem never saw); counted in the totals above",
  "sessions": [
   {
    "calls": 1,
    "dir": "-w-p",
    "first_ts": "2026-09-19T12:24:00-07:00",
    "session": "cs4",
    "src": "claude-code",
    "tokens": 110000,
    "usd": 1.5
   },
   {
    "calls": 1,
    "dir": "-w-p",
    "first_ts": "2026-09-20T05:00:00-07:00",
    "session": "orphan",
    "src": "claude-code",
    "tokens": 110,
    "usd": 0.0
   }
  ],
  "tokens": 110110,
  "usd": 1.5
 },
 "unpriced_models": [],
 "window": {
  "end_epoch_ms": 1789801200000,
  "end_exclusive_pt": "2026-09-19",
  "generated_at_pt": "2026-09-25 12:00 PT",
  "partial_last_day": false,
  "start_epoch_ms": 1789714800000,
  "start_pt": "2026-09-18"
 },
 "wins": {
  "attribution_method": null,
  "cost_status": "unmeasured",
  "dropped": [],
  "gh_unconfirmed": [],
  "items": [
   {
    "active_minutes": null,
    "confirmed": false,
    "cost_basis": null,
    "cost_label": "unmeasured (session not linked to PR)",
    "cost_status": "unmeasured",
    "day_pt": "2026-09-18",
    "evidence_ids": [
     4
    ],
    "key": "claude-mem#4125",
    "kind": "pr",
    "linked_sessions": [],
    "project": "claude-mem",
    "sessions": [
     "cs1"
    ],
    "sessions_n": 0,
    "sources": [
     "ship_observation"
    ],
    "title": "PR #4125 merged to main",
    "tokens": null,
    "ts_pt": "2026-09-18T09:30-0700",
    "turns_n": 0,
    "url": null,
    "usd": null,
    "win_id": "W-1"
   }
  ],
  "note": "finished line items (shipped/completed) are outcomes, not wins (G12); praise detection is a Phase 2B hook",
  "praise_n": 0,
  "total_attributed_usd": null,
  "unattributed_usd": null
 }
}
deprecation-and-migrationaddyosmani/agent-skillsManages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when migrating a database schema in production, such as renaming or dropping a column without downtime (expand/contract). Use when deciding whether to maintain or sunset existing code.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
performance-optimizationaddyosmani/agent-skillsOptimizes application performance across frontend, backend, queries, and databases. Use when performance requirements exist, when you suspect performance regressions, when Core Web Vitals or load times need improvement, when N+1 query patterns need fixing, or when profiling reveals bottlenecks.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
backup-setupthedotmack/claude-memSet up, check, or restore local claude-mem database snapshots. Use when the user says "set up backups", "back up my memories", "backup status", or "restore my database".Stars: 118kUpdated Claude CodeCodexVerifiedInstall