read-only demo

Videos Fu45geO3zX8

Agents Need Receipts, Not More Tool Calls - Armanas Povilionis, Alithea Bio

index_state ready data_status ok

AI Engineer· published 2026-07-18· 0:19:36· en-US· indexed 2026-08-11 00:52

Open on YouTube

Scene timeline

  1. Shot 0, 0:00 to 0:26, 1 of 1 keyframes kept
  2. Shot 1, 0:26 to 0:53, 0 of 1 keyframes kept
  3. Shot 2, 0:53 to 1:20, 0 of 1 keyframes kept
  4. Shot 3, 1:20 to 1:47, 0 of 1 keyframes kept
  5. Shot 4, 1:47 to 2:34, 1 of 1 keyframes kept
  6. Shot 5, 2:34 to 2:42, 1 of 1 keyframes kept
  7. Shot 6, 2:42 to 3:05, 1 of 1 keyframes kept
  8. Shot 7, 3:05 to 3:34, 1 of 1 keyframes kept
  9. Shot 8, 3:34 to 4:02, 0 of 1 keyframes kept
  10. Shot 9, 4:02 to 4:32, 1 of 1 keyframes kept
  11. Shot 10, 4:32 to 5:01, 0 of 1 keyframes kept
  12. Shot 11, 5:01 to 5:50, 1 of 1 keyframes kept
  13. Shot 12, 5:50 to 6:00, 1 of 1 keyframes kept
  14. Shot 13, 6:00 to 6:07, 1 of 1 keyframes kept
  15. Shot 14, 6:07 to 6:38, 1 of 1 keyframes kept
  16. Shot 15, 6:38 to 7:09, 0 of 1 keyframes kept
  17. Shot 16, 7:09 to 7:11, 1 of 1 keyframes kept
  18. Shot 17, 7:11 to 7:28, 1 of 1 keyframes kept
  19. Shot 18, 7:28 to 7:30, 1 of 1 keyframes kept
  20. Shot 19, 7:30 to 7:31, 1 of 1 keyframes kept
  21. Shot 20, 7:31 to 7:32, 0 of 1 keyframes kept
  22. Shot 21, 7:32 to 7:39, 1 of 1 keyframes kept
  23. Shot 22, 7:39 to 8:08, 1 of 1 keyframes kept
  24. Shot 23, 8:08 to 8:27, 0 of 1 keyframes kept
  25. Shot 24, 8:27 to 8:54, 0 of 1 keyframes kept
  26. Shot 25, 8:54 to 9:28, 1 of 1 keyframes kept
  27. Shot 26, 9:28 to 9:45, 0 of 1 keyframes kept
  28. Shot 27, 9:45 to 10:14, 0 of 1 keyframes kept
  29. Shot 28, 10:14 to 10:42, 0 of 1 keyframes kept
  30. Shot 29, 10:42 to 11:22, 1 of 1 keyframes kept
  31. Shot 30, 11:22 to 11:25, 1 of 1 keyframes kept
  32. Shot 31, 11:25 to 11:33, 1 of 1 keyframes kept
  33. Shot 32, 11:33 to 11:35, 1 of 1 keyframes kept
  34. Shot 33, 11:35 to 11:46, 1 of 1 keyframes kept
  35. Shot 34, 11:46 to 11:54, 1 of 1 keyframes kept
  36. Shot 35, 11:54 to 12:01, 1 of 1 keyframes kept
  37. Shot 36, 12:01 to 12:20, 1 of 1 keyframes kept
  38. Shot 37, 12:20 to 12:21, 1 of 1 keyframes kept
  39. Shot 38, 12:21 to 12:49, 1 of 1 keyframes kept
  40. Shot 39, 12:49 to 12:52, 1 of 1 keyframes kept
  41. Shot 40, 12:52 to 12:58, 1 of 1 keyframes kept
  42. Shot 41, 12:58 to 13:11, 1 of 1 keyframes kept
  43. Shot 42, 13:11 to 13:12, 1 of 1 keyframes kept
  44. Shot 43, 13:12 to 13:44, 1 of 1 keyframes kept
  45. Shot 44, 13:44 to 14:05, 1 of 1 keyframes kept
  46. Shot 45, 14:05 to 14:16, 1 of 1 keyframes kept
  47. Shot 46, 14:16 to 14:24, 1 of 1 keyframes kept
  48. Shot 47, 14:24 to 14:27, 1 of 1 keyframes kept
  49. Shot 48, 14:27 to 14:48, 1 of 1 keyframes kept
  50. Shot 49, 14:48 to 15:02, 1 of 1 keyframes kept
  51. Shot 50, 15:02 to 15:06, 1 of 1 keyframes kept
  52. Shot 51, 15:06 to 15:10, 1 of 1 keyframes kept
  53. Shot 52, 15:10 to 15:41, 0 of 1 keyframes kept
  54. Shot 53, 15:41 to 16:12, 1 of 1 keyframes kept
  55. Shot 54, 16:12 to 16:17, 1 of 1 keyframes kept
  56. Shot 55, 16:17 to 16:23, 1 of 1 keyframes kept
  57. Shot 56, 16:23 to 16:24, 1 of 1 keyframes kept
  58. Shot 57, 16:24 to 16:30, 1 of 1 keyframes kept
  59. Shot 58, 16:30 to 17:02, 1 of 1 keyframes kept
  60. Shot 59, 17:02 to 17:09, 1 of 1 keyframes kept
  61. Shot 60, 17:09 to 17:10, 1 of 1 keyframes kept
  62. Shot 61, 17:10 to 17:12, 1 of 1 keyframes kept
  63. Shot 62, 17:12 to 17:23, 1 of 1 keyframes kept
  64. Shot 63, 17:23 to 17:28, 1 of 1 keyframes kept
  65. Shot 64, 17:28 to 17:57, 0 of 1 keyframes kept
  66. Shot 65, 17:57 to 18:27, 0 of 1 keyframes kept
  67. Shot 66, 18:27 to 18:29, 0 of 1 keyframes kept
  68. Shot 67, 18:29 to 18:32, 1 of 1 keyframes kept
  69. Shot 68, 18:32 to 18:35, 0 of 1 keyframes kept
  70. Shot 69, 18:35 to 18:37, 1 of 1 keyframes kept
  71. Shot 70, 18:37 to 19:06, 1 of 1 keyframes kept
  72. Shot 71, 19:06 to 19:35, 0 of 1 keyframes kept

72 shot(s).

keyframes kept every frame deduplicated

What was stored

cues
152
whisperx 152
chunks
34
from 152 cues
keyframes
54
kept of 72 captured
frames with text
54
1,691 lines read
chapters
0
from the source metadata
keyframe bytes
8.7 MB
word timings on 152 cues

Provenance

Each pipeline stage, its state and the model that produced it
stage state model started took
fetch done 2026-08-11 00:48 1m 19s
stt done 2026-08-11 00:50 17s
chunk done 2026-08-11 00:50 0s
text_embed done 2026-08-11 00:50 1s
keyframe done 2026-08-11 00:50 1m 04s
ocr done 2026-08-11 00:51 55s
frame_embed done 2026-08-11 00:52 9s

Frames, and what the machine read

  • 0:23 #0 done3 line(s)

    shot 0·sharpness 738.8

    1. _o..o_ froglet project of Alithea bio0.93
    2. Agents need receipts,0.99
    3. not more tool calls.1.00
  • 0:32 #1 skipped

    shot 1·duplicate of #0

  • 1:04 #2 skipped

    shot 2·duplicate of #0

  • 1:33 #3 skipped

    shot 3·duplicate of #0

  • 2:24 #4 done2 line(s)

    shot 4·sharpness 714.0

    1. _0..0_ froglet0.96
    2. Alithea bio0.96
  • 2:40 #5 done2 line(s)

    shot 5·sharpness 1137.5

    1. _0..0_ froglet0.96
    2. Alithea bio0.99
  • 2:47 #6 done2 line(s)

    shot 6·sharpness 1353.5

    1. _0..0_froglet0.98
    2. Alithea bio0.99
  • 3:25 #7 done10 line(s)

    shot 7·sharpness 1323.2

    1. _0..0_ froglet0.94
    2. Alithea bio0.99
    3. AGENT SURFACES0.99
    4. PROVIDER ASSETS1.00
    5. MCP / A2A-style calls0.98
    6. data1.00
    7. OpenClaw / NemoClaw1.00
    8. compute1.00
    9. custom runtimes1.00
    10. analysis services1.00
  • 3:45 #8 skipped

    shot 8·duplicate of #7

  • 4:20 #9 done14 line(s)

    shot 9·sharpness 1674.3

    1. _0..0_ froglet0.96
    2. Alithea bio0.98
    3. AGENT INTERFACES1.00
    4. FROGLET1.00
    5. PROVIDER ASSETS1.00
    6. MCP / A2A-style calls0.98
    7. discover1.00
    8. data0.99
    9. OpenClaw / NemoClaw0.98
    10. execute1.00
    11. compute1.00
    12. custom runtimes1.00
    13. get receipt1.00
    14. analysis services1.00
  • 4:57 #10 skipped

    shot 10·duplicate of #9

  • 5:26 #11 done16 line(s)

    shot 11·sharpness 2247.9

    1. _0..0_ froglet0.94
    2. Alithea bio0.99
    3. https://froglet.dev/demo/1.00
    4. 00.97
    5. 00.96
    6. Lightning1.00
    7. Al Claude Code0.99
    8. Stripe1.00
    9. Codex1.00
    10. x+ x4020.96
    11. MCP1.00
    12. OpenClaw1.00
    13. WASM1.00
    14. 00.66
    15. Python1.00
    16. Docker1.00
  • 5:59 #12 done31 line(s)

    shot 12·sharpness 2377.2

    1. _o..o_ froglet0.90
    2. Alithea bio0.99
    3. armanas - armanas@Armanass-MacBook-Air--zsh-80×270.96
    4. FROGLET_AGENT_TARGET=manual0.99
    5. curl -fsSL https://froglet.dev/agent1.00
    6. bash1.00
    7. Installing froglet-node from1.00
    8. Installed froglet-node to /Users/armanas/.local/bin/froglet-node0.99
    9. Starting Froglet provider/runtime from published images...1.00
    10. [+] up 2/20.96
    11. Container froglet_agent-provider-1 Healthy0.99
    12. 0.5s1.00
    13. Container froglet_agent-runtime-1 Running1.00
    14. 0.0s1.00
    15. "status": "ok",0.98
    16. "bootstrap_dir": "/Users/armanas/.froglet/agent",0.99
    17. "data_dir": "/Users/armanas/.froglet/data"1.00
    18. "froglet_node": "/Users/armanas/.local/bin/froglet-node",1.00
    19. "provider_url": "http://127.0.0.1:8080",0.99
    20. "runtime_url": "http://127.0.0.1:8081",1.00
    21. "network_mode": "clearnet",1.00
    22. "compose_file": "/Users/armanas/.froglet/agent/compose.yaml",1.00
    23. "compose_project_name": "froglet_agent",1.00
    24. "compose_started": true,1.00
    25. "agent_target": "claude-code",1.00
    26. "mcp_docker_network": "froglet_agent_default",0.99
    27. "mcp_config_path": "/Users/armanas/.froglet/agent/mcp/.mcp.json",1.00
    28. "next_mcp_actions": ["status", "publish_artifact"],0.99
    29. "next_instruction": "Restart or point your agent at the MCP config path, call1.00
    30. froglet status, then publish a demo service with template demo.add."1.00
    31. < took 3s < at 18:40:550.96
  • 6:04 #13 done19 line(s)

    shot 13·sharpness 1807.2

    1. _o..o_froglet0.93
    2. Alithea bio0.97
    3. armanas- Vibe- vibe -80×240.93
    4. Vibe-vibe0.95
    5. Read https://try.froglet.dev/1lms.txt,0.99
    6. follow the hosted demo flow exactly if0.99
    7. you can access it, otnerwise say only that you could not access it, then give0.99
    8. me an honest, evidence-backed assessment that reports the observed HTTP0.99
    9. statuses, observed service IDs, observed deal status, observed result,1.00
    10. whether a receipt was present, and any mismatch between these docs and live0.99
    11. behavior before explaining what Froglet just proved, what it did not prove,1.00
    12. and the single most relevant next experiment for my files, tools, data,1.00
    13. configuration, workflows, constraints, and goals.0.99
    14. Thought0.99
    15. Fetched https://try.froglet.dev/llms.txt (11,205 chars, text/plain)1.00
    16. 114 lines0.97
    17. auto approve0.98
    18. >0.67
    19. 17% of 200k tokens0.98
  • 6:28 #14 done5 line(s)

    shot 14·sharpness 553.6

    1. _0..0_ froglet0.91
    2. Alithea bio0.96
    3. Scientific collaboration1.00
    4. setup takes years and0.99
    5. millions1.00
  • 6:59 #15 skipped

    shot 15·duplicate of #14

  • 7:10 #16 done82 line(s)

    shot 16·sharpness 2350.5

    1. Chrome1.00
    2. File1.00
    3. Edit1.00
    4. View1.00
    5. History1.00
    6. Bookmarks1.00
    7. Profiles0.96
    8. Tab1.00
    9. Window1.00
    10. Help1.00
    11. 90.78
    12. 0.73
    13. 0.65
    14. 96%0.81
    15. G0.65
    16. Fri Jun 26 18:520.99
    17. Froglet-Al-Engineer-Researc1.00
    18. x0.60
    19. Froglet- identity. execution0.96
    20. x0.79
    21. 1.00
    22. 0.94
    23. docs.google.com/presentation/d/1stTpyl3X4diA3yzawusrYeZLhXrOV73j/edit?slide=id.g3ef25a767c8_0_465#slide=id.g3ef25a767c81.00
    24. Ask Google1.00
    25. 1.00
    26. Work1.00
    27. Froglet-Al-Engineer-Research-Collaboration-v31.00
    28. .PPTX0.99
    29. 0.99
    30. 0.87
    31. Slideshow1.00
    32. Share0.95
    33. File Edit View Insert Format1.00
    34. Slide1.00
    35. Arrange Tools0.98
    36. Help1.00
    37. a + -0.60
    38. @ Fit0.95
    39. A0.56
    40. T0.71
    41. 0.94
    42. Background1.00
    43. Layout1.00
    44. Theme1.00
    45. Transition1.00
    46. ^0.69
    47. 11.00
    48. 21.00
    49. 31.00
    50. 51.00
    51. 61.00
    52. 81.00
    53. 91.00
    54. 101.00
    55. 111.00
    56. 121.00
    57. 131.00
    58. _o..0_froglet0.92
    59. AAlithea bio0.91
    60. 81.00
    61. Slide1.00
    62. Image1.00
    63. 91.00
    64. Scientific collaboration1.00
    65. setup takes years and0.99
    66. thousand tokens and takes1.00
    67. froglet setup costs few1.00
    68. Media1.00
    69. costsmillions1.00
    70. minutes.1.00
    71. 0.92
    72. 200m0.71
    73. Uploads1.00
    74. 101.00
    75. Record1.00
    76. 111.00
    77. Speaker1.00
    78. 121.00
    79. Enhance this slide ×0.97
    80. 0.89
    81. <0.97
    82. That costs few thousand tokens and take minutes1.00
  • 7:18 #17 done86 line(s)

    shot 17·sharpness 1462.4

    1. Chrome1.00
    2. File1.00
    3. Edit1.00
    4. View1.00
    5. History1.00
    6. Bookmarks1.00
    7. Profiles0.94
    8. Tab1.00
    9. Window0.94
    10. Help1.00
    11. 90.82
    12. 0.56
    13. 96%0.81
    14. G0.71
    15. Fri Jun 26 18:530.99
    16. Froglet-Al-Engineer-Researc1.00
    17. x0.67
    18. Froglet — identity. execution.0.98
    19. x0.70
    20. +0.99
    21. 1.00
    22. C0.52
    23. froglet.dev1.00
    24. 0.99
    25. Work1.00
    26. _0..0_ froglet0.95
    27. Home1.00
    28. Docs1.00
    29. Marketplace1.00
    30. Managed1.00
    31. Open source1.00
    32. 0.51
    33. GitHub0.99
    34. 40.65
    35. FROGLET - IDENTITY. EXECUTION. SETTLEMENT.0.99
    36. LET THE AGENT PROVE IT1.00
    37. Froglet is not for1.00
    38. PROMPT FOR YOUR AI AGENT0.98
    39. Open https://try.froglet.dev/llms.txt. Run0.99
    40. COPY1.00
    41. You.1.00
    42. demo.add and one witness/hash follow-up.1.00
    43. Report statuses, result, receipt/feed proof,1.00
    44. mismatches, and what this proves or does not1.00
    45. It's for your AI0.99
    46. prove for my projects/data. If unreachable,1.00
    47. say so.1.00
    48. Agent.1.00
    49. Paste into Claude Code, Codex, OpenClaw, or any agentic client. The1.00
    50. agent should come back with live evidence, not a product summary.1.00
    51. An open protocol for Al agents to sell and buy services—1.00
    52. FAST PROOF0.99
    53. STRICT AUDIT1.00
    54. zoom0.89
    55. cryptographic identity, signed deals, executable work, and verifiable1.00
    56. settlement. Trust through math, not middlemen.1.00
    57. Hosted proof0.96
    58. COPY1.00
    59. Evidence audit1.00
    60. COPY1.00
    61. FOLLOW-UP1.00
    62. LOCAL PATH0.98
    63. Try in cloud1.00
    64. Run locally0.97
    65. Watch walkthrough →0.99
    66. Proof + witness1.00
    67. Install proposal0.99
    68. COPY1.00
    69. COPW0.87
    70. FEED1.00
    71. Receipt check0.99
    72. COPY1.00
    73. AGENT REPORT1.00
    74. CATALOG1.00
    75. RESULT1.00
    76. EVIDENCE1.00
    77. No trust tour.1.00
    78. 5 free demos1.00
    79. sum = 120.99
    80. receipt + feed0.99
    81. Make the model bring back evidence from0.99
    82. demo.add + witness/hash1.00
    83. {a:7,b:5}1.00
    84. descriptor · offer0.94
    85. the live service.1.00
    86. receipt1.00
  • 7:29 #18 done82 line(s)

    shot 18·sharpness 1346.1

    1. Chrome1.00
    2. File1.00
    3. Edit1.00
    4. View1.00
    5. History1.00
    6. Bookmarks1.00
    7. Profiles1.00
    8. Tab1.00
    9. Window1.00
    10. Help1.00
    11. 90.88
    12. 0.66
    13. 96%0.83
    14. G0.65
    15. Fri Jun 26 18:530.98
    16. Froglet-Al-Engineer-Researc1.00
    17. x0.67
    18. Froglet — identity. execution.0.98
    19. x0.73
    20. +0.98
    21. 1.00
    22. 0.70
    23. froglet.dev1.00
    24. 0.99
    25. Work1.00
    26. _0..0_ froglet0.92
    27. Home1.00
    28. Docs1.00
    29. Marketplace1.00
    30. Managed1.00
    31. Open source1.00
    32. C0.80
    33. GitHub0.98
    34. You.1.00
    35. demo.add and one witness/hash follow-up.1.00
    36. Report statuses, result, receipt/feed proof,1.00
    37. mismatches, and what this proves or does not1.00
    38. It's for your AI0.99
    39. prove for my projects/data. If unreachable,1.00
    40. say so.1.00
    41. Agent.1.00
    42. Paste into Claude Code, Codex, OpenClaw, or any agentic client. The1.00
    43. agent should come back with live evidence, not a product summary.1.00
    44. An open protocol for Al agents to sell and buy services—0.99
    45. FAST PROOF1.00
    46. STRICT AUDIT1.00
    47. cryptographic identity, signed deals, executable work, and verifiable1.00
    48. Hosted proof1.00
    49. 40.94
    50. Evidence audit1.00
    51. COPY0.99
    52. COPY1.00
    53. settlement. Trust through math, not middlemen.1.00
    54. FOLLOW-UP1.00
    55. LOCAL PATH1.00
    56. Try in cloud1.00
    57. Run locally0.98
    58. Watch walkthrough →0.99
    59. Proof + witness0.99
    60. Install proposal0.98
    61. COPY0.99
    62. COPY1.00
    63. FEED0.88
    64. zoom0.84
    65. Receipt check0.99
    66. COPY0.99
    67. AGENT REPORT1.00
    68. CATALOG1.00
    69. RESULT1.00
    70. EVIDENCE1.00
    71. No trust tour.1.00
    72. 5 free demos0.99
    73. sum = 120.98
    74. receipt + feed1.00
    75. Make the model bring back evidence from0.98
    76. demo.add + witness/hash1.00
    77. {a:7,b:5}1.00
    78. descriptor · offe:0.92
    79. the live service.0.99
    80. receipt1.00
    81. LAUNCH1.00
    82. Exactly two public0.99
  • 7:30 #19 done72 line(s)

    shot 19·sharpness 1289.1

    1. Chrome1.00
    2. File1.00
    3. Edit1.00
    4. View1.00
    5. History1.00
    6. Bookmarks1.00
    7. Profiles1.00
    8. Tab1.00
    9. Window1.00
    10. Help1.00
    11. 90.88
    12. 0.89
    13. 96%0.83
    14. G0.65
    15. Fri Jun 26 18:530.96
    16. Froglet-Al-Engineer-Researc1.00
    17. x0.67
    18. Froglet — identity. execution.0.96
    19. x0.73
    20. +0.98
    21. 1.00
    22. 0.97
    23. froglet.dev1.00
    24. 1.00
    25. Work1.00
    26. _0..0_ froglet0.95
    27. Home1.00
    28. Docs1.00
    29. Marketplace1.00
    30. Managed1.00
    31. Open source1.00
    32. 0.62
    33. C0.80
    34. GitHub0.98
    35. settlement. Trust through math, not middlemen.0.99
    36. FOLLOW-UP0.98
    37. LOCAL PATH0.97
    38. Try in cloud1.00
    39. Run locally1.00
    40. Watch walkthrough →0.98
    41. Proof + witness0.99
    42. Install proposal1.00
    43. COPY1.00
    44. COPY1.00
    45. FEED0.98
    46. Receipt check0.98
    47. COPY1.00
    48. 41.00
    49. AGENT REPORT1.00
    50. CATALOG1.00
    51. RESULT1.00
    52. EVIDENCE1.00
    53. No trust tour.1.00
    54. 5 free demos0.99
    55. sum = 120.98
    56. receipt + feed0.97
    57. Make the model bring back evidence from1.00
    58. demo.add + witness/hash1.00
    59. {a:7,b:5}1.00
    60. descriptor · offer0.96
    61. the live service.1.00
    62. receipt1.00
    63. zoom0.92
    64. LAUNCH1.00
    65. Exactly two public1.00
    66. entry points.1.00
    67. Froglet doesn't apologize for constraints, it names them. Pick the one that1.00
    68. matches your intent— hosted trial or local install. No third option.0.99
    69. 01· HOSTED0.96
    70. B2 - LOCAL0.88
    71. Try in cloud0.97
    72. Run locally0.99
  • 7:32 #20 skipped

    shot 20·duplicate of #18

  • 7:34 #21 done86 line(s)

    shot 21·sharpness 1500.8

    1. Chrome1.00
    2. File1.00
    3. Edit1.00
    4. View1.00
    5. History1.00
    6. Bookmarks1.00
    7. Profiles0.94
    8. Tab1.00
    9. Window1.00
    10. Help1.00
    11. 90.82
    12. 0.63
    13. 96%0.82
    14. G0.68
    15. Fri Jun 26 18:530.98
    16. Froglet-Al-Engineer-Researc1.00
    17. x0.67
    18. Froglet — identity. execution.0.99
    19. x0.70
    20. +0.99
    21. 1.00
    22. C0.52
    23. froglet.dev1.00
    24. 0.99
    25. Work1.00
    26. _0..0_0.93
    27. froglet1.00
    28. Home1.00
    29. Docs1.00
    30. Marketplace1.00
    31. Managed1.00
    32. Open source1.00
    33. GitHub0.99
    34. FROGLET - IDENTITY. EXECUTION. SETTLEMENT.0.99
    35. LET THE AGENT PROVE IT0.99
    36. Froglet is not for1.00
    37. PROMPT FOR YOUR AI AGENT0.99
    38. Open https://try.froglet.dev/llms.txt. Run0.98
    39. COPY1.00
    40. You.1.00
    41. demo.add and one witness/hash follow-up.1.00
    42. Report statuses, result, receipt/feed proof,1.00
    43. mismatches, and what this proves or does not1.00
    44. It's for your AI1.00
    45. prove for my projects/data. If unreachable,0.99
    46. say so.1.00
    47. Agent.1.00
    48. Paste into Claude Code, Codex, OpenClaw, or any agentic client. The0.99
    49. agent should come back with live evidence, not a product summary.1.00
    50. cryptographic identity, signed deals, executable work, and verifiable1.00
    51. An open protocol for Al agents to sell and buy services—0.99
    52. FAST PROOF0.99
    53. Hosted proof1.00
    54. STRICT AUDIT1.00
    55. Evidence audit1.00
    56. COPY1.00
    57. COPY1.00
    58. zoom0.88
    59. settlement. Trust through math, not middlemen.0.99
    60. 40.93
    61. FOLLOW-UP1.00
    62. LOCAL PATH0.99
    63. Try in cloud1.00
    64. Run locally1.00
    65. Watch walkthrough →0.98
    66. Proof + witness0.97
    67. Install proposal1.00
    68. COPY1.00
    69. COPY1.00
    70. FEED1.00
    71. Receipt check0.97
    72. COPY1.00
    73. AGENT REPORT1.00
    74. CATALOG1.00
    75. RESULT1.00
    76. EVIDENCE1.00
    77. No trust tour.1.00
    78. 5 free demos0.99
    79. sum = 120.98
    80. receipt + feed1.00
    81. Make the model bring back evidence from0.98
    82. demo.add + witness/hash0.98
    83. {a:7,b:5}0.99
    84. descriptor · offer0.95
    85. the live service.1.00
    86. receipt1.00
  • 8:02 #22 done69 line(s)

    shot 22·sharpness 1225.2

    1. Chrome1.00
    2. File1.00
    3. Edit1.00
    4. View1.00
    5. History1.00
    6. Bookmarks1.00
    7. Profiles1.00
    8. Tab1.00
    9. Window1.00
    10. Help1.00
    11. 90.82
    12. 0.85
    13. 96%0.86
    14. G0.68
    15. Fri Jun 26 18:530.98
    16. Froglet-Al-Engineer-Researc1.00
    17. Demo - Froglet0.94
    18. x0.55
    19. +0.99
    20. 1.00
    21. 0.94
    22. froglet.dev/demo/1.00
    23. 0.99
    24. 0.58
    25. Work1.00
    26. _0..0_ froglet0.93
    27. Home1.00
    28. Docs1.00
    29. Marketplace1.00
    30. Managed1.00
    31. Open source1.00
    32. G0.65
    33. OGitHub0.93
    34. STEP 1 / 110.96
    35. The network1.00
    36. Three roles, one protocol. Providers sell1.00
    37. resources. Requesters buy them.1.00
    38. Marketplaces help them find each other.0.99
    39. Learn more →0.98
    40. Every node can play multiple roles. A1.00
    41. provider1.00
    42. F^P1.00
    43. deal0.89
    44. requester1.00
    45. F^R1.00
    46. marketplace is itself a provider that sells1.00
    47. search. There are no special node classes.0.99
    48. froglet - zsh0.98
    49. register1.00
    50. discover1.00
    51. zoom0.95
    52. ~ ) FROGLET_PAYMENT_BACKEND=none0.96
    53. FROGLET_PUBLISH_DEMO_SERVICES=1 cargo run --0.97
    54. bin froglet-node1.00
    55. local assumption: 127.0.0.1:8080 and :80810.99
    56. are free0.94
    57. STARTUP OUTPUT (ABRIDGED)0.99
    58. M1.00
    59. Local Runtime API: http://127.0.0.1:80810.97
    60. marketplace1.00
    61. Local API Gateway: http://127.0.0.1:80801.00
    62. Node is now online and accepting traffic.1.00
    63. ->0.53
    64. Provider Froglet → F^P0.98
    65. Requester → F^R,0.99
    66. Marketplace → M0.99
    67. 40.94
    68. Back1.00
    69. Continue1.00
  • 8:18 #23 skipped

    shot 23·duplicate of #22

Transcript

152 cues· 2,152 words· 11,811 chars

  1. 0:00 Hello, everybody.
  2. 0:02 I'm Armanos.
  3. 0:03 I work at Alifea Bio, where we focus on life sciences topics.
  4. 0:07 More specifically, we work on immuno-oncology and immunopathodontics.
  5. 0:13 While we work on science topics, we use ourselves and provide our clients agentic automation or agentic AI solutions.
  6. 0:23 However, we found a certain gap, which we would like to discuss today.
  7. 0:29 And we have a solution which we're working on, which we would like to share.
  8. 0:34 It's called Froglet.
  9. 0:37 But to start with, I would like to ask, what do you think is most valuable work that agentic automation can do?
  10. 0:49 My guess, scientific research automation would be quite high up that list.
  11. 0:56 However, from my experience, adding more tools to agents will not suffice.
  12. 1:03 because science and especially lives in especially in life sciences is inherently collaborative process and in order to enable agents to collaborate
  13. 1:18 We need a way to have a verifiable chain of receipts.
  14. 1:24 What do I mean?
  15. 1:25 We need a solution which can provide these receipts, proving every step, ensuring that every result can be trusted and enabling repeatability as well as collaboration at scale.
  16. 1:41 Having this thesis in mind, let's go to the next step.
  17. 1:45 So
  18. 1:47 Stepping back, we imagine agents as cooks in the kitchen.
  19. 1:54 Giving them more tools or better tools improves kitchen's efficiency.
  20. 2:03 Better knives, more pans, more ovens, definitely boosts the speed and quality, but it only enhances the local work.
  21. 2:14 On the other hand, scientific work is not cooking alone in your own kitchen.
  22. 2:21 It is closer to running a Michelin star restaurant.
  23. 2:26 The outcome depends on providers and quality of their produce.
  24. 2:36 high level services, as well as an ability to consistently deliver the same quality dish again and again and again.
  25. 2:45 You cannot bring everything into one kitchen.
  26. 2:50 The challenge isn't local.
  27. 2:51 It's not local tools.
  28. 2:53 The challenge is aligning the entire supply chain that it's repeatable and consistent.
  29. 3:02 Today, we have already agents with plenty of tools and increasing in volume, number, and quality.
  30. 3:12 And we also have on our hand data and specialized analytics algorithms and compute, which are distributed across different organizations and siloed.
  31. 3:26 So our vision is that whenever AI agent workflow automation matures, organizations will give to the agents not only tools, they will give them budgets.
  32. 3:43 We're already kind of doing this in a primitive form by allocating tokens for token budgets for a specific task.
  33. 3:54 But what I'm talking is a bit broader.
  34. 3:57 It's allowing agents to manage their own budgets to achieve their own goals, whether it's to discover services, request data, negotiate execution terms, or pay for work across organizational boundaries.
  35. 4:13 At that point, an agent is no longer just a cook with a better knife.
  36. 4:19 It starts to act like a chef.
  37. 4:25 like an executive chef, finding suppliers, ordering ingredients, coordinating work at the kitchen, and finally keeping the record of what has happened.
  38. 4:38 That is why we are building Froglet, an open-source protocol for agents to discover, transact with, and receive verifiable receipts from external data and service providers.
  39. 4:56 For more detailed information, obviously, go to our page, froglet.dev, where you will find general descriptions on how it works, what it integrates with.
  40. 5:09 Froglet is designed to be in between the moving parts.
  41. 5:14 It's not designed to replace.
  42. 5:15 It's designed to integrate.
  43. 5:17 It integrates with different payment rails.
  44. 5:19 It integrates with different agent harnesses, execution environments, and even transport protocols.
  45. 5:27 And it doesn't require that every node in the network uses the same software stack.
  46. 5:33 That's a beauty.
  47. 5:34 We are not forcing everyone to use the same, to work the same way.
  48. 5:38 We're just asking everybody to use the same interface.
  49. 5:44 In the webpage, you will find how to run Froglet with just, how to install it locally just with one command, or even how to run it remotely just with one prompt.
  50. 6:00 So in essence, what we're saying is closed source scientific collaboration often turns into a bespoke project, enterprise project, that can take years and cost millions before any reusable workflow exists.

Open at this second