agentsview 0.33.0__py3-none-win_arm64.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
agentsview/__init__.py ADDED
@@ -0,0 +1,10 @@
1
+ from __future__ import annotations
2
+
3
+ import subprocess
4
+ import sys
5
+ from pathlib import Path
6
+
7
+
8
+ def main() -> None:
9
+ bin_path = Path(__file__).parent / "bin" / "agentsview.exe"
10
+ sys.exit(subprocess.call([str(bin_path)] + sys.argv[1:]))
agentsview/__main__.py ADDED
@@ -0,0 +1,3 @@
1
+ from agentsview import main
2
+
3
+ main()
Binary file
@@ -0,0 +1,512 @@
1
+ Metadata-Version: 2.1
2
+ Name: agentsview
3
+ Version: 0.33.0
4
+ Summary: Local web viewer for AI agent sessions
5
+ Home-page: https://github.com/kenn-io/agentsview
6
+ Author: Kenn Software LLC
7
+ License: MIT
8
+ Requires-Python: >=3.9
9
+ Classifier: License :: OSI Approved :: MIT License
10
+ Classifier: Programming Language :: Python :: 3
11
+ Description-Content-Type: text/markdown
12
+
13
+ # agentsview
14
+
15
+ Browse, search, and track costs across all your AI coding agents. One binary, no
16
+ accounts, everything local.
17
+
18
+ <p align="center">
19
+ <img src="https://agentsview.io/screenshots/dashboard.png" alt="Analytics dashboard" width="720">
20
+ </p>
21
+
22
+ ## Install
23
+
24
+ ```bash
25
+ # macOS / Linux
26
+ curl -fsSL https://agentsview.io/install.sh | bash
27
+
28
+ # Windows
29
+ powershell -ExecutionPolicy ByPass -c "irm https://agentsview.io/install.ps1 | iex"
30
+ ```
31
+
32
+ Or download the **desktop app** (macOS / Windows) from
33
+ [GitHub Releases](https://github.com/kenn-io/agentsview/releases) or via
34
+ homebrew: `brew install --cask agentsview`
35
+
36
+ Or run the published Docker image:
37
+
38
+ ```bash
39
+ docker run --rm -p 127.0.0.1:8080:8080 \
40
+ -v agentsview-data:/data \
41
+ -v "$HOME/.claude/projects:/agents/claude:ro" \
42
+ -v "$HOME/.forge:/agents/forge:ro" \
43
+ -e CLAUDE_PROJECTS_DIR=/agents/claude \
44
+ -e FORGE_DIR=/agents/forge \
45
+ ghcr.io/kenn-io/agentsview:latest
46
+ ```
47
+
48
+ ## Quick Start
49
+
50
+ ```bash
51
+ agentsview serve # start server, open web UI
52
+ agentsview usage daily # print daily cost summary
53
+ ```
54
+
55
+ On first run, agentsview discovers sessions from every supported agent on your
56
+ machine, syncs them into a local SQLite database, and opens a web UI at
57
+ `http://127.0.0.1:8080`.
58
+
59
+ ## Remote / forwarded access
60
+
61
+ agentsview binds to loopback and validates the request `Host` header to guard
62
+ against DNS-rebinding attacks. When you reach it through SSH port-forwarding, a
63
+ reverse proxy, or a remote dev environment (exe.dev, Codespaces, Coder, WSL2),
64
+ the browser sends a `Host` that the server does not recognize, so API requests
65
+ such as `/api/v1/settings` are rejected with `403 Forbidden`.
66
+
67
+ To fix this, restart the server with `--public-url` set to the exact origin you
68
+ open in the browser:
69
+
70
+ ```bash
71
+ # Browser opens http://127.0.0.1:18080 via `ssh -L 18080:127.0.0.1:8080 host`
72
+ agentsview serve --public-url http://127.0.0.1:18080
73
+
74
+ # Browser opens a forwarded hostname
75
+ agentsview serve --public-url https://your-workspace.exe.dev
76
+ ```
77
+
78
+ Use `--public-origin` (repeatable or comma-separated) to trust additional
79
+ browser origins. If you expose the UI beyond loopback, also enable
80
+ `--require-auth`.
81
+
82
+ ## Docker
83
+
84
+ The container image defaults to local `agentsview serve`. Set `PG_SERVE=1` to
85
+ switch the startup command to `agentsview pg serve` instead.
86
+
87
+ `docker-compose.prod.yaml` is included as a production example:
88
+
89
+ ```bash
90
+ docker compose -f docker-compose.prod.yaml up -d
91
+ ```
92
+
93
+ The included compose file persists the agentsview data directory in a named
94
+ volume and mounts Claude, Codex, Forge, and OpenCode session roots read-only.
95
+ The container runs as root, so prefer a named volume for `/data` over a host
96
+ bind mount; if you do bind-mount, pre-create the directory with the desired
97
+ ownership to avoid root-owned files in your home directory.
98
+
99
+ The examples publish the UI on loopback only (`127.0.0.1`). If you need to
100
+ expose it beyond localhost, enable `--require-auth` and publish the port
101
+ intentionally.
102
+
103
+ Important: a containerized agentsview instance can only discover agent sessions
104
+ from directories you explicitly mount into the container. If you do not mount an
105
+ agent's session directory and point the matching env var at it, that agent will
106
+ not appear in the UI.
107
+
108
+ Example PostgreSQL-backed startup:
109
+
110
+ ```bash
111
+ docker run --rm -p 127.0.0.1:8080:8080 \
112
+ -e PG_SERVE=1 \
113
+ -e AGENTSVIEW_PG_URL='postgres://user:password@postgres.example.com:5432/agentsview?sslmode=require' \
114
+ ghcr.io/kenn-io/agentsview:latest
115
+ ```
116
+
117
+ Example DuckDB mirror startup:
118
+
119
+ ```bash
120
+ # Populate /data/sessions.duckdb from the mounted SQLite archive.
121
+ docker run --rm \
122
+ -v agentsview-data:/data \
123
+ -v "$HOME/.claude/projects:/agents/claude:ro" \
124
+ -e CLAUDE_PROJECTS_DIR=/agents/claude \
125
+ ghcr.io/kenn-io/agentsview:latest duckdb push --full
126
+
127
+ # Serve the populated mirror read-only.
128
+ docker run --rm -p 127.0.0.1:8080:8080 \
129
+ -v agentsview-data:/data \
130
+ ghcr.io/kenn-io/agentsview:latest duckdb serve
131
+ ```
132
+
133
+ Example Quack startup:
134
+
135
+ ```bash
136
+ # Expose the local DuckDB mirror over Quack from the host/container.
137
+ QUACK_TOKEN="$(openssl rand -base64 32)"
138
+ docker run --rm -p 127.0.0.1:9494:9494 \
139
+ -v agentsview-data:/data \
140
+ ghcr.io/kenn-io/agentsview:latest \
141
+ duckdb quack serve \
142
+ --bind quack:0.0.0.0:9494 \
143
+ --token "$QUACK_TOKEN" \
144
+ --allow-insecure
145
+
146
+ # Serve the web UI from a remote Quack endpoint.
147
+ docker run --rm -p 127.0.0.1:8080:8080 \
148
+ -e AGENTSVIEW_DUCKDB_URL='quack:https://duckdb.example.com' \
149
+ -e AGENTSVIEW_DUCKDB_TOKEN="$QUACK_TOKEN" \
150
+ ghcr.io/kenn-io/agentsview:latest duckdb serve
151
+ ```
152
+
153
+ Keep Quack on loopback or behind TLS. Plain HTTP Quack on a non-loopback bind
154
+ requires `--allow-insecure` and should only be used behind a trusted tunnel or
155
+ reverse proxy.
156
+
157
+ ## Token Usage and Cost Tracking
158
+
159
+ `agentsview usage` is a fast, local replacement for ccusage and similar tools.
160
+ It tracks token consumption and compute costs across **all** your coding agents
161
+ -- not just Claude Code. Because session data is already indexed in SQLite,
162
+ queries are over 100x faster than tools that re-parse raw session files on every
163
+ run.
164
+
165
+ ```bash
166
+ # Daily cost summary (default: last 30 days)
167
+ agentsview usage daily
168
+
169
+ # Per-model breakdown
170
+ agentsview usage daily --breakdown
171
+
172
+ # Filter by agent and date range
173
+ agentsview usage daily --agent claude --since 2026-04-01
174
+
175
+ # One-line summary for shell prompts / status bars
176
+ agentsview usage daily --all --json
177
+ agentsview usage statusline
178
+ ```
179
+
180
+ Features:
181
+
182
+ - Automatic pricing via LiteLLM rates (with offline fallback)
183
+ - Prompt-caching-aware cost calculation (cache creation / read tokens)
184
+ - Per-model breakdown with `--breakdown`
185
+ - Date filtering (`--since`, `--until`, `--all`), agent filtering (`--agent`)
186
+ - JSON output (`--json`) for scripting
187
+ - Timezone-aware date bucketing (`--timezone`)
188
+ - Works standalone -- no server required, just run the command
189
+
190
+ ## Per-Session Details
191
+
192
+ `agentsview session usage <id>` prints per-session token statistics plus a cost
193
+ estimate for a single session. The output reports the session's total output
194
+ tokens and peak context tokens, plus a cost estimate in USD (`cost_usd`) when
195
+ pricing is available for the session's model(s) (`has_cost`). Cost is computed
196
+ from input/output and cache tokens internally, but only the output-token and
197
+ peak-context totals are reported alongside the cost.
198
+
199
+ ```bash
200
+ # Print token usage and cost for a specific session
201
+ agentsview session usage <id>
202
+
203
+ # JSON output for scripting
204
+ agentsview session usage <id> --format json
205
+ ```
206
+
207
+ The same per-session usage data is available from the REST API:
208
+
209
+ ```bash
210
+ GET /api/v1/sessions/{id}/usage
211
+ ```
212
+
213
+ The response includes the `session_id`, `agent`, `project`,
214
+ `total_output_tokens`, `peak_context_tokens`, `has_token_data`, `cost_usd`,
215
+ `has_cost`, `models`, and `unpriced_models` fields from the CLI JSON schema.
216
+ HTTP responses also include `server_running: true`. Existing sessions return
217
+ `200` even when token or cost data is absent; missing sessions return `404`.
218
+
219
+ The deprecated alias `agentsview token-use <id>` remains available for
220
+ compatibility and now also reports cost estimates.
221
+
222
+ ## Session Stats
223
+
224
+ `agentsview stats` emits window-scoped analytics over recorded sessions: totals,
225
+ archetypes (automation vs. quick/standard/deep/marathon), distributions for
226
+ session duration, user-message count, peak context, and tools-per-turn, plus
227
+ cache economics, tool/model/agent mix, and a temporal hourly breakdown. The
228
+ `--format json` output follows a versioned v1 schema (`schema_version: 1`)
229
+ suitable for downstream consumers.
230
+
231
+ By default, `stats` only reads the local SQLite archive. Git-derived outcome
232
+ metrics are opt-in because they can be slow or brittle on large/missing repos:
233
+ use `--include-git-outcomes` for commits/LOC/files changed, and
234
+ `--include-github-outcomes` for GitHub PR counts via `gh` (this also enables git
235
+ outcomes).
236
+
237
+ ```bash
238
+ # Human-readable summary over the last 28 days
239
+ agentsview stats
240
+
241
+ # Machine-readable JSON over a fixed date range
242
+ agentsview stats --format json --since 2026-04-01 --until 2026-04-15
243
+
244
+ # Restrict to one agent and inspect the schema
245
+ agentsview stats --format json --agent claude | jq '.schema_version'
246
+
247
+ # Include expensive local git outcome metrics explicitly
248
+ agentsview stats --include-git-outcomes
249
+ ```
250
+
251
+ ## Session Browser
252
+
253
+ | Dashboard | Session viewer |
254
+ | ------------------------------------------------------------- | ----------------------------------------------------------------------- |
255
+ | ![Dashboard](https://agentsview.io/screenshots/dashboard.png) | ![Session viewer](https://agentsview.io/screenshots/message-viewer.png) |
256
+
257
+ | Search | Activity heatmap |
258
+ | --------------------------------------------------------------- | --------------------------------------------------------- |
259
+ | ![Search](https://agentsview.io/screenshots/search-results.png) | ![Heatmap](https://agentsview.io/screenshots/heatmap.png) |
260
+
261
+ - **Full-text search** across all message content (FTS5)
262
+ - **Token usage and cost dashboard** -- per-session and per-model cost
263
+ breakdowns, daily spend charts, all in the web UI
264
+ - **Analytics dashboard** -- activity heatmaps, tool usage, velocity metrics,
265
+ project breakdowns
266
+ - **Live updates** via SSE as active sessions receive new messages
267
+ - **Keyboard-first** navigation (`j`/`k`/`[`/`]`, `Cmd+K` search, `?` for all
268
+ shortcuts)
269
+ - **Export** sessions as HTML or publish to GitHub Gist
270
+
271
+ ## Supported Agents
272
+
273
+ agentsview auto-discovers sessions from all of these:
274
+
275
+ | Agent | Session Directory |
276
+ | ------------------ | ------------------------------------------------------ |
277
+ | Claude Code | `~/.claude/projects/` |
278
+ | Codex | `~/.codex/sessions/` |
279
+ | Copilot CLI | `~/.copilot/` |
280
+ | Gemini CLI | `~/.gemini/` |
281
+ | OpenCode | `~/.local/share/opencode/` |
282
+ | OpenHands CLI | `~/.openhands/conversations/` |
283
+ | Cursor | `~/.cursor/projects/` |
284
+ | Amp | `~/.local/share/amp/threads/` |
285
+ | iFlow | `~/.iflow/projects/` |
286
+ | Zencoder | `~/.zencoder/sessions/` |
287
+ | Zed | `~/Library/Application Support/Zed/` (macOS) |
288
+ | VSCode Copilot | `~/Library/Application Support/Code/User/` (macOS) |
289
+ | Pi | `~/.pi/agent/sessions/` |
290
+ | Qwen Code | `~/.qwen/projects/` |
291
+ | OpenClaw | `~/.openclaw/agents/` |
292
+ | QClaw | `~/.qclaw/agents/` |
293
+ | Kimi | `~/.kimi/sessions/` |
294
+ | Kiro CLI | `~/.kiro/sessions/cli/`, `~/.local/share/kiro-cli/` |
295
+ | Kiro IDE | `~/Library/Application Support/Kiro/` (macOS) |
296
+ | Cortex Code | `~/.snowflake/cortex/conversations/` |
297
+ | Hermes Agent | `~/.hermes/sessions/` |
298
+ | WorkBuddy | `~/.workbuddy/projects/` |
299
+ | Forge | `~/.forge/` |
300
+ | Piebald | `~/.local/share/piebald/` |
301
+ | Warp | `~/.warp/` (platform-dependent) |
302
+ | Positron Assistant | `~/Library/Application Support/Positron/User/` (macOS) |
303
+ | Antigravity | `~/.gemini/antigravity/` |
304
+ | Antigravity CLI | `~/.gemini/antigravity-cli/` (see note below) |
305
+
306
+ Each directory can be overridden with an environment variable. See the
307
+ [configuration docs](https://agentsview.io/configuration/) for details.
308
+
309
+ ### Antigravity CLI: high-resolution transcripts
310
+
311
+ Antigravity CLI sessions now appear in two on-disk formats. Newer releases store
312
+ conversation trajectories as SQLite `.db` files, which agentsview indexes
313
+ directly. Older releases stored assistant turns and tool calls in
314
+ AES-GCM-encrypted `.pb` files; for those sessions, agentsview falls back to
315
+ **summary mode** using your prompts from `history.jsonl` plus any plain-text
316
+ artifacts under `brain/` (plans, walkthroughs, checkpoints).
317
+
318
+ To unlock full transcripts for older `.pb` sessions, run
319
+ [agy-reader](https://github.com/mjacobs/agy-reader) alongside agentsview.
320
+ agy-reader talks to the local Antigravity daemon, decrypts each conversation,
321
+ and writes a `<uuid>.trajectory.json` sidecar next to the encrypted `.pb` file.
322
+ agentsview's file watcher detects the sidecar automatically and parses it in
323
+ place of summary mode -- no agentsview restart needed.
324
+
325
+ ```bash
326
+ go install github.com/mjacobs/agy-reader@latest
327
+
328
+ # Generate sidecars for existing sessions...
329
+ agy-reader --sync
330
+
331
+ # ...or keep them fresh as you work.
332
+ agy-reader --watch
333
+ ```
334
+
335
+ agy-reader auto-discovers the Antigravity daemon URL by parsing
336
+ `~/.gemini/antigravity-cli/cli.log`. If discovery fails (e.g. the log has
337
+ rotated), the command prints platform-specific instructions for locating the
338
+ port and exporting `ANTIGRAVITY_DAEMON_URL` manually.
339
+
340
+ Sidecars stay on your machine. agentsview makes no outbound request to produce
341
+ or read them, and treats sidecars as untrusted structured input -- see
342
+ [SECURITY.md](SECURITY.md) for the trust model.
343
+
344
+ ## PostgreSQL Sync
345
+
346
+ Push session data to a shared PostgreSQL instance for team dashboards:
347
+
348
+ ```bash
349
+ agentsview pg push # push local data to PG
350
+ agentsview pg serve # serve web UI from PG (read-only)
351
+ ```
352
+
353
+ ### Automatic push (background service)
354
+
355
+ To keep a shared PostgreSQL database current without running `pg push` by hand,
356
+ run the auto-push daemon. It watches your session directories and pushes shortly
357
+ after new sessions are recorded, with a periodic floor as a safety net:
358
+
359
+ ```bash
360
+ agentsview pg push --watch # foreground, Ctrl-C to stop
361
+ agentsview pg push --watch --debounce 1m # custom coalesce window
362
+ agentsview pg push --watch --interval 5m # custom floor interval
363
+ ```
364
+
365
+ The daemon reads the same `[pg]` config as `pg push`, so the PostgreSQL DSN must
366
+ be set in your config file (or an environment variable it expands). Protect the
367
+ config file, since it holds credentials:
368
+
369
+ ```bash
370
+ chmod 600 ~/.agentsview/config.toml
371
+ ```
372
+
373
+ To run it unattended as an OS service (launchd on macOS, `systemd --user` on
374
+ Linux):
375
+
376
+ ```bash
377
+ agentsview pg service install # generate the unit, enable + start it
378
+ agentsview pg service status # show manager status
379
+ agentsview pg service logs -f # follow the service log
380
+ agentsview pg service uninstall # stop and remove
381
+ ```
382
+
383
+ **Linux headless machines:** systemd `--user` services stop at logout and do not
384
+ start at boot unless lingering is enabled for your user. `install` detects this
385
+ and prints the command; you can also run it yourself:
386
+
387
+ ```bash
388
+ loginctl enable-linger "$USER"
389
+ ```
390
+
391
+ See [PostgreSQL docs](https://agentsview.io/postgresql/) for setup and
392
+ configuration.
393
+
394
+ ## DuckDB Mirror and Quack
395
+
396
+ DuckDB support is a mirror backend, not a replacement for the local SQLite
397
+ archive. `agentsview serve` still performs primary ingestion into SQLite. Use
398
+ DuckDB when you want a portable analytics file, read-only local serving from a
399
+ mirror, or remote read access through DuckDB's Quack protocol.
400
+
401
+ ```bash
402
+ agentsview duckdb push # mirror SQLite into DuckDB
403
+ agentsview duckdb status # show mirror sync status
404
+ agentsview duckdb serve # serve web UI from DuckDB (read-only)
405
+ agentsview duckdb quack serve # expose the local DuckDB file over Quack
406
+ ```
407
+
408
+ `agentsview duckdb serve` reads `[duckdb].path` or `AGENTSVIEW_DUCKDB_PATH`. To
409
+ serve from a remote Quack endpoint, set `AGENTSVIEW_DUCKDB_URL` and
410
+ `AGENTSVIEW_DUCKDB_TOKEN` instead. Quack is still a new DuckDB protocol, so
411
+ agentsview keeps conservative defaults: local Quack serving binds to loopback,
412
+ requires a token, and rejects non-loopback plain HTTP unless `--allow-insecure`
413
+ is explicit. For remote use, prefer a TLS URL or put Quack behind an
414
+ authenticated tunnel/proxy.
415
+
416
+ Backend modes:
417
+
418
+ - SQLite: primary local archive, file sync, FTS5 search, and writable UI.
419
+ - PostgreSQL: optional shared team backend; push from SQLite, serve read-only.
420
+ - DuckDB: optional mirror file or Quack endpoint; push from SQLite, serve
421
+ read-only.
422
+
423
+ Troubleshooting:
424
+
425
+ - If `duckdb push` fails to open the mirror, confirm the binary was built with
426
+ the DuckDB Go driver for your platform and that `AGENTSVIEW_DUCKDB_PATH`
427
+ points to a writable file location.
428
+ - If Quack commands fail with extension errors, update the agentsview binary so
429
+ the embedded DuckDB runtime includes the Quack extension.
430
+ - If a remote attach fails, check the token, the `quack:` URL, TLS/proxy
431
+ termination, and whether the server was intentionally started with
432
+ `--allow-insecure` for plain non-loopback binds.
433
+ - DuckDB search currently uses substring/regex fallback behavior. SQLite FTS5
434
+ remains the indexed search path for primary local serving.
435
+
436
+ ## Privacy
437
+
438
+ agentsview sends a limited anonymous `daemon_active` telemetry ping to PostHog
439
+ when the server starts and every 24 hours while it runs, using a stable random
440
+ install ID as the event `DistinctId`. The event includes
441
+ `application=agentsview`, app version, commit, OS, and CPU architecture, with
442
+ `$process_person_profile=false` and `$geoip_disable=true`. It does not include
443
+ session, project, prompt, file path, account, or machine identity. Disable
444
+ telemetry with `AGENTSVIEW_TELEMETRY_ENABLED=0` or `TELEMETRY_ENABLED=0`.
445
+ Telemetry is also hard-disabled in Go test binaries, regardless of environment.
446
+
447
+ All session data stays on your machine. The server binds to `127.0.0.1` by
448
+ default. The update check is optional and can be disabled with
449
+ `--no-update-check`.
450
+
451
+ ## Documentation
452
+
453
+ Full docs at **[agentsview.io](https://agentsview.io)**:
454
+ [Quick Start](https://agentsview.io/quickstart/) --
455
+ [Usage Guide](https://agentsview.io/usage/) --
456
+ [CLI Reference](https://agentsview.io/commands/) --
457
+ [Configuration](https://agentsview.io/configuration/) --
458
+ [Architecture](https://agentsview.io/architecture/)
459
+
460
+ ______________________________________________________________________
461
+
462
+ ## Development
463
+
464
+ Requires Go 1.26+ (CGO), Node.js 22+.
465
+
466
+ ```bash
467
+ make dev # Go server (dev mode)
468
+ make frontend-dev # Vite dev server (run alongside make dev)
469
+ make build # build binary with embedded frontend
470
+ make install # install to ~/.local/bin
471
+ ```
472
+
473
+ ```bash
474
+ make test # Go tests (CGO_ENABLED=1 -tags "fts5,kit_posthog_disabled")
475
+ make bench-backends # compare SQLite, DuckDB, and PostgreSQL store reads
476
+ make lint # golangci-lint + NilAway
477
+ make nilaway # NilAway through custom golangci-lint
478
+ make e2e # Playwright E2E tests
479
+ ```
480
+
481
+ `make bench-backends` requires Docker. It starts a PostgreSQL container with
482
+ testcontainers, mirrors the same SQLite fixture into DuckDB and PostgreSQL, and
483
+ benchmarks the shared `db.Store` read queries for relative comparison. The
484
+ default fixture is 1,000 sessions and 64,000 messages; use
485
+ `BENCH_BACKENDS_SESSIONS` and `BENCH_BACKENDS_MESSAGES_PER_SESSION` to scale it.
486
+ When the Docker CLI uses a non-default socket, export `DOCKER_HOST` for that
487
+ socket before running the benchmark.
488
+
489
+ Pre-commit hooks via [prek](https://github.com/j178/prek): run `make lint-tools`
490
+ and `make install-hooks` after cloning (requires `prek` and `uv`).
491
+
492
+ ### Project Layout
493
+
494
+ ```
495
+ cmd/agentsview/ CLI entrypoint
496
+ internal/ Go packages (config, db, parser, server, sync, postgres)
497
+ frontend/ Svelte 5 SPA (Vite, TypeScript)
498
+ desktop/ Tauri desktop wrapper
499
+ ```
500
+
501
+ ## Acknowledgements
502
+
503
+ Inspired by
504
+ [claude-history-tool](https://github.com/andyfischer/ai-coding-tools/tree/main/claude-history-tool)
505
+ by Andy Fischer and
506
+ [claude-code-transcripts](https://github.com/simonw/claude-code-transcripts) by
507
+ Simon Willison.
508
+
509
+ ## License
510
+
511
+ MIT
512
+
@@ -0,0 +1,7 @@
1
+ agentsview/__init__.py,sha256=BjbaBtM3mOH2liyrj2Rt19UPJH2uJRAl8JMBhJQBVM0,238
2
+ agentsview/__main__.py,sha256=TmPx60CQc_hxduT618L0HffKfbfGKtSNUW9gHhxTBsY,36
3
+ agentsview/bin/agentsview.exe,sha256=w550Xz9v0-8sKpyMSEziCC9A0NnqHYHcw40TjIX7Igg,27148288
4
+ agentsview-0.33.0.dist-info/METADATA,sha256=C63xU8ZJYJuyYcOTnwVfFTwCbqETiP80UUywPUNzyoQ,20675
5
+ agentsview-0.33.0.dist-info/WHEEL,sha256=5YJ5xtW4Ie1I6n82yBVxATGBwb8xfCzl0s8Lhw_xanI,101
6
+ agentsview-0.33.0.dist-info/entry_points.txt,sha256=4-mB9InmBn_2B4RAcxFu_3wqpNfXgOuFq1jvVn0Mg3U,47
7
+ agentsview-0.33.0.dist-info/RECORD,,
@@ -0,0 +1,4 @@
1
+ Wheel-Version: 1.0
2
+ Generator: agentsview-build-wheels
3
+ Root-Is-Purelib: false
4
+ Tag: py3-none-win_arm64
@@ -0,0 +1,2 @@
1
+ [console_scripts]
2
+ agentsview = agentsview:main